DETAILED ACTION
1. The present application is being examined under the pre-AIA first to invent provisions.
Continued Examination Under 37 CFR 1.114
2. A request for continued examination under 37 CFR 1.114, including the fee set forth in 37 CFR 1.17(e), was filed in this application after final rejection. Since this application is eligible for continued examination under 37 CFR 1.114, and the fee set forth in 37 CFR 1.17(e) has been timely paid, the finality of the previous Office action has been withdrawn pursuant to 37 CFR 1.114. Applicant's submission filed on 03/20/2026 has been entered.
Response to Amendment
3. Receipt of Applicant’s Amendment filed on 03/20/2026 is acknowledged. The amendment includes the amending of claims 1, 9, 11, 15, 17, 19, and 28-30.
Terminal Disclaimer
4. The terminal disclaimers filed on 05/12/2022 disclaiming the terminal portion of any patent granted on this application which would extend beyond the expiration date of U.S. Patent 10,515,120 and U.S. Patent 10,902,064 have been reviewed and are accepted. The terminal disclaimers have been recorded.
Claim Rejections - 35 USC § 101
5. 35 U.S.C. 101 reads as follows:
Whoever invents or discovers any new and useful process, machine, manufacture, or composition of matter, or any new and useful improvement thereof, may obtain a patent therefor, subject to the conditions and requirements of this title.
6. Claims 1-5, 7-12, 14-17, 19-23, 27-28, and 29 are rejected under 35 U.S.C. 101 because the claimed invention is directed to non-statutory subject matter because the claimed invention is directed to a judicial exception (i.e., a law of nature, a natural phenomenon, or an abstract idea) without significantly more.
Under the 2019 PEG, when considering subject matter eligibility under 35 U.S.C. § 101, it must be determined whether the claim is directed to one of the four statutory categories of invention, i.e., process, machine, manufacture, or composition of matter (step 1). If the claim does fall within one of the statutory categories, it must then be determined whether the claim is directed to a judicial exception (i.e., law of nature, natural phenomenon, and abstract idea) (step 2A prong 1), and if so, it must additionally be determined whether the claim is integrated into a practical application (step 2A prong 2). If an abstract idea is present in the claim without integration into a practical application, any element or combination of elements in the claim must be sufficient to ensure that the claim amounts to significantly more than the abstract idea itself (step 2B).
In the instant case, claims (1-5, 7-10, and 19-21), (11-12, 14-16, 22, and 27), (17, 23, and 28), and (29) are directed to a computer-implemented method, system, non-transitory computer readable medium, and computer-implemented method respectively. Thus, each of the claims falls within one of the four statutory categories. However, the claims also fall within the judicial exception of an abstract idea.
Under Step 2A Prong 1, the test is to identify whether the claims are “directed to” a judicial exception. The examiner notes that the claimed invention is directed to an abstract idea in that the instant application is directed to mental processes, specifically recommending an item based on current access of a user.
The examiner further notes that claims 1, 11, 17, and 29 recite a computer-implemented method, system, non-transitory computer readable medium, and computer-implemented method for recommending an item which is similar to themes defined above of method of mental processes such as recommending an item, and is similar to the abstract idea identified in the 2019 PEG in grouping “c” in that the claims recite certain methods of mental processes such as performing the recommending of an item based on a current access of a user. The limitations, substantially comprising the body of the claim, recite a standard process of determining a recommendation of an item. The examiner notes that the claimed invention recommends an item based on a current access of a user. Because the limitations above closely follow the steps in recommending an item, and the steps of the claims involve mental processes, the claim recites an abstract idea consistent with the “mental processes” grouping set forth in the 2019 PEG.
Claim 1:
A computer-implemented method, comprising: receiving data from a data storage device associated with a catalog of items and a particular item;
calculating, by digital electronic circuitry, similarity scores between the particular item and other items in the catalog;
wherein the calculating comprises: identifying terms associated with the particular item and the other items;
generating a vector for the particular item and the other items;
wherein the vector represents a frequency score for each term associated with the particular item and the other items; and
normalizing each vector;
retrieving, from data stores, an access history and associations associated with the particular item and the other items in the catalog;
providing the access history and associations associated with the particular item and the other items in the catalog to a predictive model;
wherein the predictive model comprises a second-order rank of associations;
determining, by the digital electronic circuitry, weights of the predictive model based on the access histories and the associations for the particular item and the other items in view of the similarity scores;
detecting a current access of the particular item;
providing data regarding access of the particular item to the predictive model; and
outputting data associated with one of the other items based on an output of the predictive model; and
transmitting the one of the other items over a computer network for viewing by a user at a network access point;
wherein the associations comprise artifacts collected by a content management system that relate certain items to information about the user.
These limitations, as drafted, is an apparatus that, under its broadest reasonable interpretation, covers the performance of mental processes specifically recommending an item based on a current access of a user. Recommending an item has long before the modern computer was invented, and continues to be predominantly a product of human endeavor. The instant application is directed to recommending an item based off of a current access of a user. Additionally, the claimed calculation of similarity scores can be performed by a human via their mind and/or pen & paper. Furthermore, the claimed identification of terms associated with the particular item and other items can be performed by a human via their mind and/or pen & paper. Moreover, the claimed generating of vectors (that represent a frequency value) can be performed by a human via their mind and/or pen & paper. Additionally, the claimed normalizing of such vectors can be performed by a human via their mind and/or pen & paper. Furthermore, the claimed providing of an access history and associations to a model that comprises a second-order rank of associations can be performed by a human via their mind and/or pen & paper. Moreover, the claimed determination of weights can be performed by a human via their mind and/or pen & paper. Additionally, the claimed determination of a current access can be performed by a human via their mind and/or pen & paper. Furthermore, the claimed providing of a current access to a model can be performed by a human via their mind and/or pen & paper. Moreover, the claimed defined artifacts is simply defining data that can be performed by a human via their mind and/or pen & paper.
The mere nominal recitation of generic computer components such as a data storage device, digital electronic circuitry, data stores, a computer network, a network access point, and a content management system do not take the claim out of the certain methods of organizing human activity grouping.
Because the limitations above closely follow the steps of recommending an item, and the steps involved human judgments, observations and evaluations that can be practically or reasonably performed in the human mind and/or pen & paper, the claim recites an abstract idea consistent with the “mental process” grouping set forth in the 2019 PEG.
If the claims are directed toward the judicial exception of an abstract idea, it must then be determined under Step 2A Prong 2 whether the judicial exception is integrated into a practical application. Examiner notes that considerations under Step 2A Prong 2 comprise most the consideration previously evaluated in the context of Step 2B. The Examiner submits that the considerations discussed previously determined that the claim does not recite “significantly more” at Step 2B would be evaluated the same under Step 2A Prong 1 and result in the determination that the claim does not integrate the abstract idea into a practical application.
The instant application fails to integrate the judicial exception into a practical application because the instant application merely recites words “apply it” (or an equivalent) with the judicial exception or merely includes instructions to implement an abstract idea. The instant application is directed to an apparatus instructing the reader to implement the identified apparatus of mental processes of recommending an item based off of a current access of a user. The elements of the claim do not themselves amount to an improvement to the computer, to a technology or another technical field. Moreover, receiving of data from a data storage device is a data gathering operation that is an insignificant data gathering operation that does not integrate the abstract idea into a practical application. Furthermore, the retrieving of an access history and associations from data stores is a data gathering operation that is an insignificant data gathering operation that does not integrate the abstract idea into a practical application. Moreover, the outputting of data is a data outputting operation that is an insignificant data outputting operation that does not integrate the abstract idea into a practical application. Additionally, the transmission of the one of the other items over a network to a network access point is simply a data transmission operation that is an insignificant data transmission operation that does not integrate the abstract idea into a practical application (See also Section 2106.05(d)(II) of the MPEP).
Here, the instructions entirely comprise the abstract idea, leaving little if any aspects of the claim for further consideration under Step 2A Prong 2. In short, the role of the generic computing elements recited in claim 1 is the same as the role of the computer in the claims considered by the Supreme Court in Alice, and the claim as whole amounts merely to an instruction to apply the abstract idea on the generic computing system/platform. Therefore, the claims have failed to integrate a practical application (see at least 84 Fed. Reg. (4) at 55). Under the 2019 PEG, this supports the conclusion that the claim is directed to an abstract idea, and the analysis proceeds to Step 2B.
While many considerations in Step 2A need not be reevaluated in Step 2B because the outcome will be the same. Here, on the basis of the additional elements other than the abstract idea, considered individually and in combination as discussed above, the Examiner respectfully submits that the claim 1 does not contain any additional elements that individually or as an ordered combination amount to an inventive concept and the claims are ineligible.
Indeed, receiving of data from a data storage device and retrieving access history are data gathering operations that are insignificant extra-solution activity that does not amount to significantly more than the abstract idea. Furthermore, outputting data is an insignificant extra-solution activity that does not amount to significantly more than the abstract idea. Moreover, transmitting data over a network is also deemed to be an insignificant extra-solution activity (See Revised Guidance 55, n.31, see also MPEP 2106.05(d)(II)(i) & MPEP 2106.05(g)).
With respect to the dependent claims do not recite anything that is found to render the abstract idea as being transformed into a patent eligible invention. The dependent claims are merely reciting further embellishments of the abstract idea and do not claim anything that amounts to significantly more than the abstract idea itself.
With respect to the dependent claims, they have been considered and are not found to be reciting anything that amounts to being significantly more than the abstract idea. Claims 2-5, 7-10, and 19-21 are directed to further embellishments of the central theme of the abstract idea in that the claims are directed to further embellishments of the recommending an item based off of current access of a user of the steps of claim 1 and do not amount to significantly more.
Specifically, claim 2 is simply further defining the abstract idea of recommending an item based off of a user’s current access of the steps of claim 1 by defining the time metric and can be performed by the human mind and/or pen & paper and does not amount to significantly more.
Moreover, claim 3 is simply further defining the abstract idea of recommending an item based off of a user’s current access of the steps of claim 1 by defining the time metric and can be performed by the human mind and/or pen & paper and does not amount to significantly more.
Furthermore, claim 4 is simply further defining the abstract idea of recommending an item based off of a user’s current access of the steps of claim 1 by defining the time metric and can be performed by the human mind and/or pen & paper and does not amount to significantly more.
Additionally, claim 5 is simply further defining the abstract idea of recommending an item based off of a user’s current access of the steps of claim 1 by defining the time metric and can be performed by the human mind and/or pen & paper and does not amount to significantly more.
Furthermore, claim 7 is simply further defining the abstract idea of recommending an item based off of a user’s current access of the steps of claim 1 by defining the interaction metric and can be performed by the human mind and/or pen & paper and does not amount to significantly more.
Additionally, claim 8 is simply further defining the abstract idea of recommending an item based off of a user’s current access of the steps of claim 1 by defining a calculation of a similarity metric and can be performed by the human mind and/or pen & paper and does not amount to significantly more.
Moreover, claim 9 is simply further defining the abstract idea of recommending an item based off of a user’s current access of the steps of claim 1 by defining an output and can be performed by the human mind and/or pen & paper and does not amount to significantly more. Specifically, defining an output is merely an insignificant extra-solution activity and does not amount to significantly more.
Furthermore, claim 10 is simply further defining the abstract idea of recommending an item based off of a user’s current access of the steps of claim 1 by defining the product that is to be licensed/have permissions and can be performed by the human mind and/or pen & paper and does not amount to significantly more.
Moreover, claim 19 is simply further defining the abstract idea of recommending an item based off of a user’s current access of the steps of claim 1 by defining an output and can be performed by the human mind and/or pen & paper and does not amount to significantly more. Specifically, defining an output is merely an insignificant extra-solution activity and does not amount to significantly more.
Furthermore, claim 20 is simply further defining the abstract idea of recommending an item based off of a user’s current access of the steps of claim 1 by defining the predicted preference based off of calculated metrics and can be performed by the human mind and/or pen & paper and does not amount to significantly more.
Additionally, claim 21 is simply further defining the abstract idea of recommending an item based off of a user’s current access of the steps of claim 1 by defining the defining the type of data that is obtained and can be performed by the human mind and/or pen & paper and does not amount to significantly more. Specifically, defining data that is obtained is merely an insignificant extra-solution activity as it is a mere data gathering operation and does not amount to significantly more.
Claim 11:
A computer-implemented system, comprising: one or more data processors;
a non-transitory computer-readable medium encoded with instructions for commanding the one or more data processors to execute steps of a method that include: receiving data from a storage device associated with a catalog of items and a particular item;
calculating similarity scores between the particular item and other items in the catalog;
wherein the calculating comprises: identifying terms associated with the particular item and the other items;
generating a vector for the particular item and the other items;
wherein the vector represents a frequency score for each term associated with the particular item and the other items; and
normalizing each vector;
retrieving, from data stores, an access history and associations associated with the particular item and the other items in the catalog;
providing the access history and associations associated with the particular item and the other items in the catalog to a predictive model;
wherein the predictive model comprises a second-order rank of associations;
determining, by the digital electronic circuitry, weights of the predictive model based on the access histories and the associations for the particular item and the other items in view of the similarity scores;
detecting a current access of the particular item;
providing data regarding access of the particular item to the predictive model;
outputting data associated with one of the other items based on an output of the predictive model; and
transmitting the one of the other items over a computer network for viewing by a user at a network access point;
wherein the associations comprise artifacts collected by a content management system that relate certain items to information about the user.
These limitations, as drafted, is an apparatus that, under its broadest reasonable interpretation, covers the performance of mental processes specifically recommending an item based on a current access of a user. Recommending an item has long before the modern computer was invented, and continues to be predominantly a product of human endeavor. The instant application is directed to recommending an item based off of a current access of a user. Additionally, the claimed calculation of similarity scores can be performed by a human via their mind and/or pen & paper. Furthermore, the claimed identification of terms associated with the particular item and other items can be performed by a human via their mind and/or pen & paper. Moreover, the claimed generating of vectors (that represent a frequency value) can be performed by a human via their mind and/or pen & paper. Additionally, the claimed normalizing of such vectors can be performed by a human via their mind and/or pen & paper. Furthermore, the claimed providing of an access history and associations to a model that comprises a second-order rank of associations can be performed by a human via their mind and/or pen & paper. Moreover, the claimed determination of weights can be performed by a human via their mind and/or pen & paper. Additionally, the claimed determination of a current access can be performed by a human via their mind and/or pen & paper. Furthermore, the claimed providing of a current access to a model can be performed by a human via their mind and/or pen & paper. Moreover, the claimed defined artifacts is simply defining data that can be performed by a human via their mind and/or pen & paper.
The mere nominal recitation of generic computer components such as one or more data processors, a non-transitory computer-readable medium, a data storage device, data stores, a computer network, a network access point, and a content management system do not take the claim out of the certain methods of organizing human activity grouping.
Because the limitations above closely follow the steps of recommending an item, and the steps involved human judgments, observations and evaluations that can be practically or reasonably performed in the human mind and/or pen & paper, the claim recites an abstract idea consistent with the “mental process” grouping set forth in the 2019 PEG.
If the claims are directed toward the judicial exception of an abstract idea, it must then be determined under Step 2A Prong 2 whether the judicial exception is integrated into a practical application. Examiner notes that considerations under Step 2A Prong 2 comprise most the consideration previously evaluated in the context of Step 2B. The Examiner submits that the considerations discussed previously determined that the claim does not recite “significantly more” at Step 2B would be evaluated the same under Step 2A Prong 1 and result in the determination that the claim does not integrate the abstract idea into a practical application.
The instant application fails to integrate the judicial exception into a practical application because the instant application merely recites words “apply it” (or an equivalent) with the judicial exception or merely includes instructions to implement an abstract idea. The instant application is directed to an apparatus instructing the reader to implement the identified apparatus of mental processes of recommending an item based off of a current access of a user. The elements of the claim do not themselves amount to an improvement to the computer, to a technology or another technical field. Moreover, receiving of data from a data storage device is a data gathering operation that is an insignificant data gathering operation that does not integrate the abstract idea into a practical application. Furthermore, the retrieving of an access history and associations from data stores is a data gathering operation that is an insignificant data gathering operation that does not integrate the abstract idea into a practical application. Moreover, the outputting of data is a data outputting operation that is an insignificant data outputting operation that does not integrate the abstract idea into a practical application. Additionally, the transmission of the one of the other items over a network to a network access point is simply a data transmission operation that is an insignificant data transmission operation that does not integrate the abstract idea into a practical application (See also Section 2106.05(d)(II) of the MPEP).
Here, the instructions entirely comprise the abstract idea, leaving little if any aspects of the claim for further consideration under Step 2A Prong 2. In short, the role of the generic computing elements recited in claim 1 is the same as the role of the computer in the claims considered by the Supreme Court in Alice, and the claim as whole amounts merely to an instruction to apply the abstract idea on the generic computing system/platform. Therefore, the claims have failed to integrate a practical application (see at least 84 Fed. Reg. (4) at 55). Under the 2019 PEG, this supports the conclusion that the claim is directed to an abstract idea, and the analysis proceeds to Step 2B.
While many considerations in Step 2A need not be reevaluated in Step 2B because the outcome will be the same. Here, on the basis of the additional elements other than the abstract idea, considered individually and in combination as discussed above, the Examiner respectfully submits that the claim 11 does not contain any additional elements that individually or as an ordered combination amount to an inventive concept and the claims are ineligible.
Indeed, receiving of data from a data storage device and retrieving access history are data gathering operations that are insignificant extra-solution activity that does not amount to significantly more than the abstract idea. Furthermore, outputting data is an insignificant extra-solution activity that does not amount to significantly more than the abstract idea. Moreover, transmitting data over a network is also deemed to be an insignificant extra-solution activity (See Revised Guidance 55, n.31, see also MPEP 2106.05(d)(II)(i) & MPEP 2106.05(g)).
With respect to the dependent claims, they have been considered and are not found to be reciting anything that amounts to being significantly more than the abstract idea. Claims 12, 14-16, 22, and 27 are directed to further embellishments of the central theme of the abstract idea in that the claims are directed to further embellishments of the performing the legal duties of recommending an item based off of a user’s current access of the steps of claim 11 and do not amount to significantly more.
Specifically, claim 12 is simply further defining the abstract idea of recommending an item based off of a user’s current access of the steps of claim 11 by defining the time metric and can be performed by the human mind and/or pen & paper and does not amount to significantly more.
Furthermore, claim 14 is simply further defining the abstract idea of recommending an item based off of a user’s current access of the steps of claim 11 by defining a similarity score and can be performed by the human mind and/or pen & paper and does not amount to significantly more.
Additionally claim 15 is simply further defining the abstract idea of recommending an item based off of a user’s current access of the steps of claim 1 by defining an output and can be performed by the human mind and/or pen & paper and does not amount to significantly more. Specifically, defining an output is merely an insignificant extra-solution activity and does not amount to significantly more.
Moreover, claim 16 is simply further defining the abstract idea of recommending an item based off of a user’s current access of the steps of claim 11 by defining the product that is to be licensed/have permissions and can be performed by the human mind and/or pen & paper and does not amount to significantly more.
Furthermore, claim 22 is simply further defining the abstract idea of recommending an item based off of a user’s current access of the steps of claim 1 by defining the defining the type of data that is obtained and can be performed by the human mind and/or pen & paper and does not amount to significantly more. Specifically, defining data that is obtained is merely an insignificant extra-solution activity as it is a mere data gathering operation and does not amount to significantly more.
Additionally claim 27 is simply further defining the abstract idea of recommending an item based off of a user’s current access of the steps of claim 11 by defining the interaction metric and can be performed by the human mind and/or pen & paper and does not amount to significantly more.
Claim 17:
A non-transitory computer-readable medium encoded with instructions for commanding one or more data processors to execute the steps of a method, the steps comprising: receiving data from a storage device associated with a catalog of items and a particular item;
calculating, by digital electronic circuitry, similarity scores between the particular item and other items in the catalog;
wherein the calculating comprises: identifying terms associated with the particular item and the other items;
generating a vector for the particular item and the other items;
wherein the vector represents a frequency score for each term associated with the particular item and the other items; and
normalizing each vector;
retrieving, from data stores, an access history and associations associated with the particular item and the other items in the catalog;
providing the access history and associations associated with the particular item and the other items in the catalog to a predictive model;
wherein the predictive model comprises a second-order rank of associations;
determining, by the digital electronic circuitry, weights of the predictive model based on the access histories and the associations for the particular item and the other items in view of the similarity scores;
detecting a current access of the particular item;
providing data regarding access of the particular item to the predictive model;
outputting data associated with one of the other items based on an output of the predictive model;
transmitting the one of the other items over a computer network for viewing by a user at a network access point;
wherein the associations comprise artifacts collected by a content management system that relate certain items to information about the user.
These limitations, as drafted, is an apparatus that, under its broadest reasonable interpretation, covers the performance of mental processes specifically recommending an item based on a current access of a user. Recommending an item has long before the modern computer was invented, and continues to be predominantly a product of human endeavor. The instant application is directed to recommending an item based off of a current access of a user. Additionally, the claimed calculation of similarity scores can be performed by a human via their mind and/or pen & paper. Furthermore, the claimed identification of terms associated with the particular item and other items can be performed by a human via their mind and/or pen & paper. Moreover, the claimed generating of vectors (that represent a frequency value) can be performed by a human via their mind and/or pen & paper. Additionally, the claimed normalizing of such vectors can be performed by a human via their mind and/or pen & paper. Furthermore, the claimed providing of an access history and associations to a model that comprises a second-order rank of associations can be performed by a human via their mind and/or pen & paper. Moreover, the claimed determination of weights can be performed by a human via their mind and/or pen & paper. Additionally, the claimed determination of a current access can be performed by a human via their mind and/or pen & paper. Furthermore, the claimed providing of a current access to a model can be performed by a human via their mind and/or pen & paper. Moreover, the claimed defined artifacts is simply defining data that can be performed by a human via their mind and/or pen & paper.
The mere nominal recitation of generic computer components such as one or more data processors, a data storage device, digital electronic circuitry, data stores, a computer network, a network access point, and a content management system do not take the claim out of the certain methods of organizing human activity grouping.
Because the limitations above closely follow the steps of recommending an item, and the steps involved human judgments, observations and evaluations that can be practically or reasonably performed in the human mind and/or pen & paper, the claim recites an abstract idea consistent with the “mental process” grouping set forth in the 2019 PEG.
If the claims are directed toward the judicial exception of an abstract idea, it must then be determined under Step 2A Prong 2 whether the judicial exception is integrated into a practical application. Examiner notes that considerations under Step 2A Prong 2 comprise most the consideration previously evaluated in the context of Step 2B. The Examiner submits that the considerations discussed previously determined that the claim does not recite “significantly more” at Step 2B would be evaluated the same under Step 2A Prong 1 and result in the determination that the claim does not integrate the abstract idea into a practical application.
The instant application fails to integrate the judicial exception into a practical application because the instant application merely recites words “apply it” (or an equivalent) with the judicial exception or merely includes instructions to implement an abstract idea. The instant application is directed to an apparatus instructing the reader to implement the identified apparatus of mental processes of recommending an item based off of a current access of a user. The elements of the claim do not themselves amount to an improvement to the computer, to a technology or another technical field. Moreover, receiving of data from a data storage device is a data gathering operation that is an insignificant data gathering operation that does not integrate the abstract idea into a practical application. Furthermore, the retrieving of an access history and associations from data stores is a data gathering operation that is an insignificant data gathering operation that does not integrate the abstract idea into a practical application. Moreover, the outputting of data is a data outputting operation that is an insignificant data outputting operation that does not integrate the abstract idea into a practical application. Additionally, the transmission of the one of the other items over a network to a network access point is simply a data transmission operation that is an insignificant data transmission operation that does not integrate the abstract idea into a practical application (See also Section 2106.05(d)(II) of the MPEP).
Here, the instructions entirely comprise the abstract idea, leaving little if any aspects of the claim for further consideration under Step 2A Prong 2. In short, the role of the generic computing elements recited in claim 1 is the same as the role of the computer in the claims considered by the Supreme Court in Alice, and the claim as whole amounts merely to an instruction to apply the abstract idea on the generic computing system/platform. Therefore, the claims have failed to integrate a practical application (see at least 84 Fed. Reg. (4) at 55). Under the 2019 PEG, this supports the conclusion that the claim is directed to an abstract idea, and the analysis proceeds to Step 2B.
While many considerations in Step 2A need not be reevaluated in Step 2B because the outcome will be the same. Here, on the basis of the additional elements other than the abstract idea, considered individually and in combination as discussed above, the Examiner respectfully submits that the claim 17 does not contain any additional elements that individually or as an ordered combination amount to an inventive concept and the claims are ineligible.
Indeed, receiving of data from a data storage device and retrieving access history are data gathering operations that are insignificant extra-solution activity that does not amount to significantly more than the abstract idea. Furthermore, outputting data is an insignificant extra-solution activity that does not amount to significantly more than the abstract idea. Moreover, transmitting data over a network is also deemed to be an insignificant extra-solution activity (See Revised Guidance 55, n.31, see also MPEP 2106.05(d)(II)(i) & MPEP 2106.05(g)).
With respect to the dependent claims, they have been considered and are not found to be reciting anything that amounts to being significantly more than the abstract idea. Claims 23 and 28 are directed to further embellishments of the central theme of the abstract idea in that the claims are directed to further embellishments of the performing the legal duties of recommending an item based off of a user’s current access of the steps of claim 17 and do not amount to significantly more.
Specifically, claim 22 is simply further defining the abstract idea of recommending an item based off of a user’s current access of the steps of claim 1 by defining the defining the type of data that is obtained and can be performed by the human mind and/or pen & paper and does not amount to significantly more. Specifically, defining data that is obtained is merely an insignificant extra-solution activity as it is a mere data gathering operation and does not amount to significantly more.
Additionally claim 28 is simply further defining the abstract idea of recommending an item based off of a user’s current access of the steps of claim 17 by defining the interaction metric and can be performed by the human mind and/or pen & paper and does not amount to significantly more.
Claim 29:
A computer-implemented method comprising: receiving data from a data storage device associated with a catalog of items and a particular item;
calculating, by digital electronic circuitry, similarity scores between the particular item and other items in the catalog;
wherein the calculating comprises: identifying terms associated with the particular item and the other items;
generating a vector for the particular item and the other items;
wherein the vector represents a frequency score for each term associated with the particular item and the other items; and
normalizing each vector;
retrieving, from data stores, an access history and associations associated with the particular item and the other items in the catalog;
providing the access history and the associations associated with the particular item and the other items in the catalog to a predictive model;
wherein the predictive model comprises a second-order rank of associations;
determining, by the digital electronic circuitry, weights of the predictive model based on the access histories and the associations for the particular item and the other items in view of the similarity scores;
outputting data associated with one of the other items based on an output of the predictive model; and
transmitting the one of the other items over a computer network for viewing by a user at a network access point;
wherein the associations comprise artifacts collected by a content management system that relate to certain items to information about the user.
These limitations, as drafted, is an apparatus that, under its broadest reasonable interpretation, covers the performance of mental processes specifically recommending an item based on a current access of a user. Recommending an item has long before the modern computer was invented, and continues to be predominantly a product of human endeavor. The instant application is directed to recommending an item based off of a current access of a user. Additionally, the claimed calculation of similarity scores can be performed by a human via their mind and/or pen & paper. Furthermore, the claimed generating of vectors (that represent a frequency value) can be performed by a human via their mind and/or pen & paper. Moreover, the claimed normalizing of such vectors can be performed by a human via their mind and/or pen & paper. Additionally, the claimed providing of an access history and associations to a model that comprises a second-order rank of associations can be performed by a human via their mind and/or pen & paper. Furthermore, the claimed determination of weights can be performed by a human via their mind and/or pen & paper. Moreover, the claimed defined artifacts is simply defining data that can be performed by a human via their mind and/or pen & paper.
The mere nominal recitation of generic computer components such as a data storage device, digital electronic circuitry, data stores, a computer network, a network access point, and a content management system do not take the claim out of the certain methods of organizing human activity grouping.
Because the limitations above closely follow the steps of recommending an item, and the steps involved human judgments, observations and evaluations that can be practically or reasonably performed in the human mind and/or pen & paper, the claim recites an abstract idea consistent with the “mental process” grouping set forth in the 2019 PEG.
If the claims are directed toward the judicial exception of an abstract idea, it must then be determined under Step 2A Prong 2 whether the judicial exception is integrated into a practical application. Examiner notes that considerations under Step 2A Prong 2 comprise most the consideration previously evaluated in the context of Step 2B. The Examiner submits that the considerations discussed previously determined that the claim does not recite “significantly more” at Step 2B would be evaluated the same under Step 2A Prong 1 and result in the determination that the claim does not integrate the abstract idea into a practical application.
The instant application fails to integrate the judicial exception into a practical application because the instant application merely recites words “apply it” (or an equivalent) with the judicial exception or merely includes instructions to implement an abstract idea. The instant application is directed to an apparatus instructing the reader to implement the identified apparatus of mental processes of recommending an item based off of a current access of a user. The elements of the claim do not themselves amount to an improvement to the computer, to a technology or another technical field. Moreover, receiving of data from a data storage device is a data gathering operation that is an insignificant data gathering operation that does not integrate the abstract idea into a practical application. Furthermore, the retrieving of an access history and associations is a data gathering operation that is an insignificant data gathering operation that does not integrate the abstract idea into a practical application. Moreover, the outputting of data is a data outputting operation that is an insignificant data outputting operation that does not integrate the abstract idea into a practical application. Additionally, the transmission of the one of the other items over a network to a network access point is simply a data transmission operation that is an insignificant data transmission operation that does not integrate the abstract idea into a practical application (See also Section 2106.05(d)(II) of the MPEP).
Here, the instructions entirely comprise the abstract idea, leaving little if any aspects of the claim for further consideration under Step 2A Prong 2. In short, the role of the generic computing elements recited in claim 1 is the same as the role of the computer in the claims considered by the Supreme Court in Alice, and the claim as whole amounts merely to an instruction to apply the abstract idea on the generic computing system/platform. Therefore, the claims have failed to integrate a practical application (see at least 84 Fed. Reg. (4) at 55). Under the 2019 PEG, this supports the conclusion that the claim is directed to an abstract idea, and the analysis proceeds to Step 2B.
While many considerations in Step 2A need not be reevaluated in Step 2B because the outcome will be the same. Here, on the basis of the additional elements other than the abstract idea, considered individually and in combination as discussed above, the Examiner respectfully submits that the claim 17 does not contain any additional elements that individually or as an ordered combination amount to an inventive concept and the claims are ineligible.
Indeed, receiving of data from a data storage device and retrieving access history are data gathering operations that are insignificant extra-solution activity that does not amount to significantly more than the abstract idea. Furthermore, outputting data is an insignificant extra-solution activity that does not amount to significantly more than the abstract idea. Moreover, transmitting data over a network is also deemed to be an insignificant extra-solution activity (See Revised Guidance 55, n.31, see also MPEP 2106.05(d)(II)(i) & MPEP 2106.05(g)).
6. Under the 2019 PEG, when considering subject matter eligibility under 35 U.S.C. § 101, it must be determined whether the claim is directed to one of the four statutory categories of invention, i.e., process, machine, manufacture, or composition of matter (step 1). If the claim does fall within one of the statutory categories, it must then be determined whether the claim is directed to a judicial exception (i.e., law of nature, natural phenomenon, and abstract idea) (step 2A prong 1), and if so, it must additionally be determined whether the claim is integrated into a practical application (step 2A prong 2). If an abstract idea is present in the claim without integration into a practical application, any element or combination of elements in the claim must be sufficient to ensure that the claim amounts to significantly more than the abstract idea itself (step 2B).
In the instant case, claims 30 and 33-34 are directed to a computer-implemented method. Thus, each of the claims falls within one of the four statutory categories. However, the claims also fall within the judicial exception of an abstract idea.
Under Step 2A Prong 1, the test is to identify whether the claims are “directed to” a judicial exception. The examiner notes that the claimed invention is directed to an abstract idea in that the instant application is directed to mental processes, specifically recommending an item based on current access of a user.
The examiner further notes that claim 30 recites a computer-implemented method for identifying a trendsetter which is similar to themes defined above of method of mental processes such as recommending an item, and is similar to the abstract idea identified in the 2019 PEG in grouping “c” in that the claims recite certain methods of mental processes such as performing the recommending of an item based on a current access of a user. The limitations, substantially comprising the body of the claim, recite a standard process of identifying a trendsetter. The examiner notes that the claimed invention identifies a trendsetter. Because the limitations above closely follow the steps in identifying a trendsetter, and the steps of the claims involve mental processes, the claim recites an abstract idea consistent with the “mental processes” grouping set forth in the 2019 PEG.
Claim 30:
A computer-implemented method comprising: receiving data from a data storage device associated with a catalog of items and a particular item;
calculating, by digital electronic circuitry, similarity scores between the particular item and other items in the catalog;
wherein the calculating comprises: identifying terms associated with the particular item and the other items;
generating a vector for the particular item and the other items;
wherein the vector represents a frequency score for each term associated with the particular item and the other items; and
normalizing each vector;
retrieving, from data stores, an access history and associations associated with the particular item and the other items in the catalog;
providing the access history and the associations associated with the particular item and the other items in the catalog to a predictive model;
wherein the predictive model comprises a second-order rank of associations;
determining, by the digital electronic circuitry, weights of the predictive model based on the access histories and the associations for the particular item and the other items in view of the similarity scores;
identifying a particular person from a pool for the particular item based on an output of the predictive model indicating that the person was someone who interacted early with a similar, popular item;
wherein the associations comprise artifacts collected by a content management system that relate to certain items to information about the user.
These limitations, as drafted, is an apparatus that, under its broadest reasonable interpretation, covers the performance of mental processes specifically identifying a trendsetter. Identifying a trendsetter has long before the modern computer was invented, and continues to be predominantly a product of human endeavor. The instant application is directed to identifying a trendsetter. Additionally, the claimed calculation of similarity scores can be performed by a human via their mind and/or pen & paper. Furthermore, the claimed identification of terms associated with the particular item and other items can be performed by a human via their mind and/or pen & paper. Moreover, the claimed generating of vectors (that represent a frequency value) can be performed by a human via their mind and/or pen & paper. Additionally, the claimed normalizing of such vectors can be performed by a human via their mind and/or pen & paper. Furthermore, the claimed providing of an access history and associations to a model that comprises a second-order rank of associations can be performed by a human via their mind and/or pen & paper. Moreover, the claimed determination of weights can be performed by a human via their mind and/or pen & paper. Additionally, the claimed identifying of a specific person from a pool of persons can be performed by a human via their mind and/or pen & paper. Furthermore, the claimed defined artifacts is simply defining data that can be performed by a human via their mind and/or pen & paper.
The mere nominal recitation of generic computer components such as a data storage device, digital electronic circuitry, data stores, and a content management system do not take the claim out of the certain methods of organizing human activity grouping.
Because the limitations above closely follow the steps of recommending an item, and the steps involved human judgments, observations and evaluations that can be practically or reasonably performed in the human mind and/or pen & paper, the claim recites an abstract idea consistent with the “mental process” grouping set forth in the 2019 PEG.
If the claims are directed toward the judicial exception of an abstract idea, it must then be determined under Step 2A Prong 2 whether the judicial exception is integrated into a practical application. Examiner notes that considerations under Step 2A Prong 2 comprise most the consideration previously evaluated in the context of Step 2B. The Examiner submits that the considerations discussed previously determined that the claim does not recite “significantly more” at Step 2B would be evaluated the same under Step 2A Prong 1 and result in the determination that the claim does not integrate the abstract idea into a practical application.
The instant application fails to integrate the judicial exception into a practical application because the instant application merely recites words “apply it” (or an equivalent) with the judicial exception or merely includes instructions to implement an abstract idea. The instant application is directed to an apparatus instructing the reader to implement the identified apparatus of mental processes of recommending an item based off of a current access of a user. The elements of the claim do not themselves amount to an improvement to the computer, to a technology or another technical field. Moreover, receiving of data from a data storage device is a data gathering operation that is an insignificant data gathering operation that does not integrate the abstract idea into a practical application. Furthermore, the retrieving of an access history and associations is a data gathering operation that is an insignificant data gathering operation that does not integrate the abstract idea into a practical application. Moreover, the outputting of data is a data outputting operation that is an insignificant data outputting operation that does not integrate the abstract idea into a practical application.
Here, the instructions entirely comprise the abstract idea, leaving little if any aspects of the claim for further consideration under Step 2A Prong 2. In short, the role of the generic computing elements recited in claim 1 is the same as the role of the computer in the claims considered by the Supreme Court in Alice, and the claim as whole amounts merely to an instruction to apply the abstract idea on the generic computing system/platform. Therefore, the claims have failed to integrate a practical application (see at least 84 Fed. Reg. (4) at 55). Under the 2019 PEG, this supports the conclusion that the claim is directed to an abstract idea, and the analysis proceeds to Step 2B.
While many considerations in Step 2A need not be reevaluated in Step 2B because the outcome will be the same. Here, on the basis of the additional elements other than the abstract idea, considered individually and in combination as discussed above, the Examiner respectfully submits that the claim 30 does not contain any additional elements that individually or as an ordered combination amount to an inventive concept and the claims are ineligible.
Indeed, receiving of data from a data storage device and retrieving access history are data gathering operations that are insignificant extra-solution activity that does not amount to significantly more than the abstract idea. Furthermore, outputting data is an insignificant extra-solution activity that does not amount to significantly more than the abstract idea.
With respect to the dependent claims, they have been considered and are not found to be reciting anything that amounts to being significantly more than the abstract idea. Claims 33-34 are directed to further embellishments of the central theme of the abstract idea in that the claims are directed to further embellishments of the recommending an item based off of current access of a user of the steps of claim 30 and do not amount to significantly more.
Specifically, claim 33 is simply further defining the abstract idea of identifying a trendsetter of the steps of claim 30 by defining the interaction metric and can be performed by the human mind and/or pen & paper and does not amount to significantly more.
Furthermore, claim 34 is directed towards the defining of what the similarity scores are based on and which can be performed by the human mind and/or pen & paper and does not amount to significantly more.
Claim Rejections - 35 USC § 103
7. In the event the determination of the status of the application as subject to AIA 35 U.S.C. 102 and 103 (or as subject to pre-AIA 35 U.S.C. 102 and 103) is incorrect, any correction of the statutory basis for the rejection will not be considered a new ground of rejection if the prior art relied upon, and the rationale supporting the rejection, would be the same under either status.
8. The following is a quotation of pre-AIA 35 U.S.C. 103(a) which forms the basis for all obviousness rejections set forth in this Office action:
(a) A patent may not be obtained though the invention is not identically disclosed or described as set forth in section 102, if the differences between the subject matter sought to be patented and the prior art are such that the subject matter as a whole would have been obvious at the time the invention was made to a person having ordinary skill in the art to which said subject matter pertains. Patentability shall not be negatived by the manner in which the invention was made.
9. Claims 1-2, 4-5, 8, 10-12, 15-17, 19-23, and 27-29 are rejected under pre-AIA 35 U.S.C. 103(a) as being unpatentable over of Dicker et al. (U.S. PGPUB 2010/0191619) in view of Hassanzadeh et al. (Article entitled “Linked Movie Data Base”, dated 20 April 2009), and further in view of Chand et al. (U.S. PGPUB 2007/0276826).
10. Regarding claims 1, 11, and 17, Dicker teaches a computer-implemented method, computer-implemented system, and non-transitory computer-readable medium comprising:
A) receiving data from a data storage device associated with a catalog of items and a particular item (Paragraph 58);
B) calculating, by digital electronic circuitry, similarity scores between the particular item and other items in the catalog (Paragraph 76, Figure 1);
G) retrieving, from data stores, an access history and associations associated with the particular item and the other items in the catalog (Paragraphs 58, 74, 129-135, and 209);
H) providing the access history and associations associated with the particular item and the other items in the catalog to a predictive model (Paragraphs 74, 97, 99, and 128-135);
J) determining, by the digital electronic circuitry, weights of the predictive model based on the access histories and the associations for the particular item and the other items in view of the similarity scores (Paragraphs 85 and 128-135);
K) detecting a current access of the particular item (Paragraphs 193 and 195, Figure 12);
L) providing data regarding access of the particular item to the predictive model (Paragraphs 193 and 195, Figure 12); and
M) outputting data associated with one of the other items based on an output of the predictive model (Paragraphs 193 and 195, Figure 12);
N) transmitting the one of the other items over a computer network for viewing by a user at a network access point (Paragraphs 193 and 195, Figure 12);
O) wherein the associations comprise artifacts collected by a content management system that relate certain items to information about the user (Paragraphs 58 and 85).
The examiner notes that Dicker teaches “receiving data from a data storage device associated with a catalog of items and a particular item” as “The Web site 30 also includes a "user profiles" database 38 which stores account-specific information about users of the site. Because a group of individuals can share an account, a given "user" from the perspective of the Web site may include multiple actual users. As illustrated by FIG. 1, the data stored for each user may include one or more of the following types of information (among other things) that can be used to generate recommendations: (a) the user's purchase history, including dates of purchase, (b) a history of items recently viewed by the user, (c) the user's item ratings profile (if any), (d) the current contents of the user's personal shopping cart(s), and (e) a listing of items that were recently (e.g., within the last six months) removed from the shopping cart(s) without being purchased ("recent shopping cart contents")” (Paragraph 58). The examiner further notes that purchase history, viewing history, ratings, etc. teach the undefined claimed data. The examiner further notes that Dicker teaches “calculating, by digital electronic circuitry, similarity scores between the particular item and other items in the catalog” as “Each similar items list 64 consists of the N (e.g., 20) items which, based on correlations between purchases of items, are deemed to be the most closely related to the respective popular item 62. Each item in the similar items list 64 is stored together with a commonality index ("CI") value which indicates the relatedness of that item to the popular item 62, based on sales of the respective items. A relatively high commonality index for a pair of items ITEM A and ITEM B indicates that a relatively large percentage of users who bought ITEM A also bought ITEM B (and vice versa). A relatively low commonality index for ITEM A and ITEM B indicates that a relatively small percentage of the users who bought ITEM A also bought ITEM B (and vice versa). As described below, the similar items lists are generated, for each popular item, by selecting the N other items that have the highest commonality index values” (Paragraph 76). The examiner further notes that computing commonality index values between each popular item 62 (i.e. the claimed particular item) and other items (see similar items list 64 in Figure 1) teaches the claimed calculating. The examiner further notes that Dicker teaches “retrieving, from data stores, an access history and associations associated with the particular item and the other items in the catalog” as “The Web site 30 also includes a "user profiles" database 38 which stores account-specific information about users of the site. Because a group of individuals can share an account, a given "user" from the perspective of the Web site may include multiple actual users. As illustrated by FIG. 1, the data stored for each user may include one or more of the following types of information (among other things) that can be used to generate recommendations: (a) the user's purchase history, including dates of purchase, (b) a history of items recently viewed by the user, (c) the user's item ratings profile (if any), (d) the current contents of the user's personal shopping cart(s), and (e) a listing of items that were recently (e.g., within the last six months) removed from the shopping cart(s) without being purchased ("recent shopping cart contents")” (Paragraph 58), “Each entry in the similar items table 60 is preferably in the form of a mapping of a popular item 62 to a corresponding list 64 of similar items ("similar items lists"). As used herein, a "popular" item is an item which satisfies some pre-specified popularity criteria. For example, in the embodiment described herein, an item is treated as popular of it has been purchased by more than 30 customers during the life of the Web site. Using this criteria produces a set of popular items (and thus a recommendation service) which grows over time. The similar items list 64 for a given popular item 62 may include other popular items” (Paragraph 74), “The process flows shown in FIGS. 3A and 3B differ primarily in that they use different types of user actions as evidence of users' interests in a particular items. In the method shown in FIG. 3A, a user is assumed to be interested in an item if the user purchased the item; and in the process shown in 3B, a user is assumed to be interested in an item if the user viewed the item. Any of a variety of other types of user actions that evidence a user's interest in a particular item may additionally or alternatively be used, alone or in combination, to generate the similar items table 60. The following are examples of other types of user actions that may be used for this purpose:… (4) Submitting a favorable review for an item. With this method, products A and B may be treated as related if a large percentage of those favorably reviewed A also favorably reviewed B. A favorable review may be defined as a score that satisfies a particular threshold (e.g., 4 or above on a scale of 1-5)… With the above and other types of item-affinity-evidencing actions, equation (1) above may be used to generate the CI values” (Paragraphs 129-135), and “In step 650 the user's profile, or a portion of the profile, is read from a user database 38 (FIG. 1) or cache” (Paragraph 209). The examiner further notes that ascertaining popularity of the commercial items entails determining a purchase history (i.e. an “access history” in the broadest reasonable interpretation) of those commercial items. Moreover, retrieved user profile information includes various of amounts of data (such as ratings profile data (i.e. the claimed associations in the broadest reasonable interpretation (See also Page 6, lines 10-15 of the instant specification that defines the claimed associations as including liked items from a user))) that can be retrieved from multiple stores such as a database and/or cache. The examiner further notes that Dicker teaches “providing the access history and associations associated with the particular item and the other items in the catalog to a predictive model” as “Each entry in the similar items table 60 is preferably in the form of a mapping of a popular item 62 to a corresponding list 64 of similar items ("similar items lists"). As used herein, a "popular" item is an item which satisfies some pre-specified popularity criteria. For example, in the embodiment described herein, an item is treated as popular of it has been purchased by more than 30 customers during the life of the Web site. Using this criteria produces a set of popular items (and thus a recommendation service) which grows over time. The similar items list 64 for a given popular item 62 may include other popular items” (Paragraph 74), “In step 106, the process identifies the items that constitute "popular" items. This may be accomplished, for example, by selecting from the item-to-customers table 104A those items that were purchased by more than a threshold number (e.g., 30) of customers. In the context of a merchant Web site such as that of Amazon.com, Inc., the resulting set of popular items may contain hundreds of thousands or millions of items” (Paragraph 97), “In step 110, the process generates the commonality indexes for each (popular_item, other_item) pair in the table 108A. As indicated above, the commonality index (CI) values are measures of the similarity between two items, with larger CI values indicating greater degrees of similarity” (Paragraph 99), “The process flows shown in FIGS. 3A and 3B differ primarily in that they use different types of user actions as evidence of users' interests in a particular items. In the method shown in FIG. 3A, a user is assumed to be interested in an item if the user purchased the item; and in the process shown in 3B, a user is assumed to be interested in an item if the user viewed the item. Any of a variety of other types of user actions that evidence a user's interest in a particular item may additionally or alternatively be used, alone or in combination, to generate the similar items table 60. The following are examples of other types of user actions that may be used for this purpose:… (4) Submitting a favorable review for an item. With this method, products A and B may be treated as related if a large percentage of those favorably reviewed A also favorably reviewed B. A favorable review may be defined as a score that satisfies a particular threshold (e.g., 4 or above on a scale of 1-5)… With the above and other types of item-affinity-evidencing actions, equation (1) above may be used to generate the CI values” (Paragraphs 129-135). The examiner further notes that providing purchase history data (i.e. access history) and/or review data (i.e. the claimed associations in the broadest reasonable interpretation) to the recommendation system of Dicker (i.e. a “predictive model”) results in the calculation of CI values for subsequent recommendations. The examiner further notes that Dicker teaches “determining, by the digital electronic circuitry, weights of the predictive model based on the access histories and the associations for the particular item and the other items in view of the similarity scores” as “In step 84, the similar items lists 64 are optionally weighted based on information about the user's affinity for the corresponding items of known interest. For example, a similar items list 64 may be weighted heavily if the user gave the corresponding popular item a rating of "5" on a scale or 1-5, or if the user purchased multiple copies of the item. Weighting a similar items list 64 heavily has the effect of increasing the likelihood that the items in that list we be included in the recommendations ultimately presented to the user. In one implementation described below, the user is presumed to have a greater affinity for recently purchased items over earlier purchased items. Similarly, where viewing histories are used to identify items of interest, items viewed recently may be weighted more heavily than earlier viewed items” (Paragraph 85) and “The process flows shown in FIGS. 3A and 3B differ primarily in that they use different types of user actions as evidence of users' interests in a particular items. In the method shown in FIG. 3A, a user is assumed to be interested in an item if the user purchased the item; and in the process shown in 3B, a user is assumed to be interested in an item if the user viewed the item. Any of a variety of other types of user actions that evidence a user's interest in a particular item may additionally or alternatively be used, alone or in combination, to generate the similar items table 60. The following are examples of other types of user actions that may be used for this purpose:… (4) Submitting a favorable review for an item. With this method, products A and B may be treated as related if a large percentage of those favorably reviewed A also favorably reviewed B. A favorable review may be defined as a score that satisfies a particular threshold (e.g., 4 or above on a scale of 1-5)… With the above and other types of item-affinity-evidencing actions, equation (1) above may be used to generate the CI values” (Paragraphs 129-135). The examiner further notes that a particular item and other items of a similar items list 64 are weighted via weights based off of user affinity data (that can include access histories and/or ratings data (i.e. an example of the claimed associations in the broadest reasonable interpretation)). The examiner further notes that Dicker teaches “detecting a current access of the particular item” as “the table 60 may be used to display "canned" lists of related items on product detail pages of the "popular" items (i.e., items for which a similar items list 64 exists). FIG. 12 illustrates this feature in example form. In this example, the detail page of a product is supplemented with the message "customers who viewed this item also viewed the following items," followed by a hypertextual list 500 of four related items. In this particular embodiment, the list is generated from the viewing-history-based version of the similar items table (generated as described in section IV-B)” (Paragraph 193) and “FIG. 13 illustrates a process that may be used to generate a related items list 500 of the type shown in FIG. 12. As illustrated, the related items list 500 for a given product is generated by retrieving the corresponding similar items list 64 (preferably from a viewing-history-based similar items table 60 as described above), optionally filtering out items falling outside the product category of the product, and then extracting the N top-rank items” (Paragraph 195). The examiner further notes that viewing (i.e. a “current access” in the broadest reasonable interpretation) a specific product (See example of the MP3 player in Figure 12) entails detecting such an access in the first place. The examiner further notes that Dicker teaches “providing data regarding access of the particular item to the predictive model” as “the table 60 may be used to display "canned" lists of related items on product detail pages of the "popular" items (i.e., items for which a similar items list 64 exists). FIG. 12 illustrates this feature in example form. In this example, the detail page of a product is supplemented with the message "customers who viewed this item also viewed the following items," followed by a hypertextual list 500 of four related items. In this particular embodiment, the list is generated from the viewing-history-based version of the similar items table (generated as described in section IV-B)” (Paragraph 193) and “FIG. 13 illustrates a process that may be used to generate a related items list 500 of the type shown in FIG. 12. As illustrated, the related items list 500 for a given product is generated by retrieving the corresponding similar items list 64 (preferably from a viewing-history-based similar items table 60 as described above), optionally filtering out items falling outside the product category of the product, and then extracting the N top-rank items” (Paragraph 195). The examiner further notes that viewing (i.e. a “current access” in the broadest reasonable interpretation) a specific product (See example of the MP3 player in Figure 12) results in the display of recommended products via the recommendation process (i.e. the “predictive model”) of Dicker. The examiner further notes that Dicker teaches “outputting data associated with one of the other items based on an output of the predictive model” as “the table 60 may be used to display "canned" lists of related items on product detail pages of the "popular" items (i.e., items for which a similar items list 64 exists). FIG. 12 illustrates this feature in example form. In this example, the detail page of a product is supplemented with the message "customers who viewed this item also viewed the following items," followed by a hypertextual list 500 of four related items. In this particular embodiment, the list is generated from the viewing-history-based version of the similar items table (generated as described in section IV-B)” (Paragraph 193) and “FIG. 13 illustrates a process that may be used to generate a related items list 500 of the type shown in FIG. 12. As illustrated, the related items list 500 for a given product is generated by retrieving the corresponding similar items list 64 (preferably from a viewing-history-based similar items table 60 as described above), optionally filtering out items falling outside the product category of the product, and then extracting the N top-rank items” (Paragraph 195). The examiner further notes that the displayed recommended similar items (See Figure 12) of Dicker teaches the claimed output. The examiner further notes that Dicker teaches “transmitting the one of the other items over a computer network for viewing by a user at a network access point” as “the table 60 may be used to display "canned" lists of related items on product detail pages of the "popular" items (i.e., items for which a similar items list 64 exists). FIG. 12 illustrates this feature in example form. In this example, the detail page of a product is supplemented with the message "customers who viewed this item also viewed the following items," followed by a hypertextual list 500 of four related items. In this particular embodiment, the list is generated from the viewing-history-based version of the similar items table (generated as described in section IV-B)” (Paragraph 193) and “FIG. 13 illustrates a process that may be used to generate a related items list 500 of the type shown in FIG. 12. As illustrated, the related items list 500 for a given product is generated by retrieving the corresponding similar items list 64 (preferably from a viewing-history-based similar items table 60 as described above), optionally filtering out items falling outside the product category of the product, and then extracting the N top-rank items” (Paragraph 195). The examiner further notes that the displayed recommended similar items (See Figure 12) of Dicker (which shows an Amazon web page) teaches the claimed transmission over a network for viewing by a user via their computing device (i.e. a network access point). The examiner further notes that Dicker teaches “wherein the associations comprise artifacts collected by a content management system that relate certain items to information about the user” as “The Web site 30 also includes a "user profiles" database 38 which stores account-specific information about users of the site. Because a group of individuals can share an account, a given "user" from the perspective of the Web site may include multiple actual users. As illustrated by FIG. 1, the data stored for each user may include one or more of the following types of information (among other things) that can be used to generate recommendations: (a) the user's purchase history, including dates of purchase, (b) a history of items recently viewed by the user, (c) the user's item ratings profile (if any), (d) the current contents of the user's personal shopping cart(s), and (e) a listing of items that were recently (e.g., within the last six months) removed from the shopping cart(s) without being purchased ("recent shopping cart contents")” (Paragraph 58) and “In step 84, the similar items lists 64 are optionally weighted based on information about the user's affinity for the corresponding items of known interest. For example, a similar items list 64 may be weighted heavily if the user gave the corresponding popular item a rating of "5" on a scale or 1-5, or if the user purchased multiple copies of the item” (Paragraph 85). The examiner further notes that retrieved user profile information includes various of amounts of data (such as ratings profile data (i.e. the claimed associations in the broadest reasonable interpretation (See also Page 6, lines 10-15 of the instant specification that defines the claimed associations as including liked items from a user))).
Dicker does not explicitly teach:
C) wherein the calculating comprises: identifying terms associated with the particular item and the other items;
D) generating a vector for the particular item and the other items;
E) wherein the vector represents a frequency score for each term associated with the particular item and the other items;
F) normalizing each vector.
Hassanzadeh, however, teaches “wherein the calculating comprises: identifying terms associated with the particular item and the other items” as “The tf-idf cosine similarity is a well established measure in the IR community which leverages the vector space model. This measure determines the closeness of the input strings r1 and r2 by first transforming the strings into unit vectors and then measuring the angle between their corresponding vectors. The cosine similarity with tf-idf weights is given by… where wr1 (t) and wr2 (t) are the normalized tf-idf weights for each common token in r1 and r2 respectively. The normalized tf-idf weight of token t in a given string record r is defined as follows… where tfr(t) is the term frequency of token t within string r and idf(t) is the inverse document frequency with respect to the entire relation R” (Section 4.1.3) and “We matched 38,064 movie titles in our database with 25,424 movie titles from DBpedia using the similarity predicates described above” (Section 4.3.2), “generating a vector for the particular item and the other items” as “The tf-idf cosine similarity is a well established measure in the IR community which leverages the vector space model. This measure determines the closeness of the input strings r1 and r2 by first transforming the strings into unit vectors and then measuring the angle between their corresponding vectors. The cosine similarity with tf-idf weights is given by… where wr1 (t) and wr2 (t) are the normalized tf-idf weights for each common token in r1 and r2 respectively. The normalized tf-idf weight of token t in a given string record r is defined as follows… where tfr(t) is the term frequency of token t within string r and idf(t) is the inverse document frequency with respect to the entire relation R” (Section 4.1.3) and “We matched 38,064 movie titles in our database with 25,424 movie titles from DBpedia using the similarity predicates described above” (Section 4.3.2), “wherein the vector represents a frequency score for each term associated with the particular item and the other items” as “The tf-idf cosine similarity is a well established measure in the IR community which leverages the vector space model. This measure determines the closeness of the input strings r1 and r2 by first transforming the strings into unit vectors and then measuring the angle between their corresponding vectors. The cosine similarity with tf-idf weights is given by… where wr1 (t) and wr2 (t) are the normalized tf-idf weights for each common token in r1 and r2 respectively. The normalized tf-idf weight of token t in a given string record r is defined as follows… where tfr(t) is the term frequency of token t within string r and idf(t) is the inverse document frequency with respect to the entire relation R” (Section 4.1.3) and “We matched 38,064 movie titles in our database with 25,424 movie titles from DBpedia using the similarity predicates described above” (Section 4.3.2), and “normalizing each vector” as “The tf-idf cosine similarity is a well established measure in the IR community which leverages the vector space model. This measure determines the closeness of the input strings r1 and r2 by first transforming the strings into unit vectors and then measuring the angle between their corresponding vectors. The cosine similarity with tf-idf weights is given by… where wr1 (t) and wr2 (t) are the normalized tf-idf weights for each common token in r1 and r2 respectively. The normalized tf-idf weight of token t in a given string record r is defined as follows… where tfr(t) is the term frequency of token t within string r and idf(t) is the inverse document frequency with respect to the entire relation R” (Section 4.1.3) and “We matched 38,064 movie titles in our database with 25,424 movie titles from DBpedia using the similarity predicates described above” (Section 4.3.2).
The examiner further notes that the secondary reference of Hassanzadeh teaches the concept of identifying terms of items for subsequent generation of vectors (which are then normalized) representing those items. Such vectors are representative of TF-IDF. The combination would result in using such a normalized vectors in the item similarity of Dicker.
It would have been obvious to one of ordinary skill in the art at the time the invention was made to combine the teachings of the cited references because teaching Hassanzadeh’s would have allowed Dicker’s to provide a method for high accuracy and/or scalable similarity algorithms, as noted by Hassanzadeh (Section 4.1).
Dicker and Hassanzadeh do not explicitly teach:
I) wherein the predictive model comprises a second-order rank of associations.
Chand, however, teaches “wherein the predictive model comprises a second-order rank of associations” as “A method is provided to aggregate a plurality of affinity lists to generate a single aggregated affinity list representing predicted affinities of a particular item, to other items, under a plurality of conditions” (Abstract), “The inventors have discovered that, by considering both affinity and lift values, a plurality of affinity lists (which, for example, might be used to generate item-based recommendations) can be advantageously aggregated, to generate a personalized recommendation” (Paragraph 16), and “The aggregated list is sort based on the aggregated inverse distances… this final ranked-aggregated list may serve as a basis for making recommendations to Ryan of other artists in which Ryan is likely to be interested” (Paragraph 38).
The examiner further notes that the secondary reference of Chand teaches the concept of a recommendation system (i.e. a predictive model) including a second-order rank of associations (which is interpreted as simply an aggregation of affinities (See Page 7, lines 21-23 of the instant specification)). The combination would result in the predictive recommendation model of Dicker to also use the undefined claimed second-order rank of associations.
It would have been obvious to one of ordinary skill in the art at the time the invention was made to combine the teachings of the cited references because teaching Chand’s would have allowed Dicker’s and Hassanzadeh’s to provide a method for providing personalized recommendations based off of multiple user characteristics, as noted by Chand (Paragraph 16).
Regarding claims 2 and 12, Dicker further teaches a computer-implemented method and computer-implemented system comprising:
A) wherein the weights of the predictive model are further based on an amount of time interaction metric (Paragraph 147).
The examiner notes that Dicker teaches “wherein the weights of the predictive model are further based on an amount of time interaction metric” as “In step 184, the process 52 weights each similar items list based on the duration since the associated popular item was purchased by the user (with recently-purchased items weighted more heavily), or if the popular item was not purchased, the rating given to the popular item by the user” (Paragraph 147). The examiner further notes that the weights of Dicker being based on a duration metric (i.e. the claimed amount of time interaction metric in the broadest reasonable interpretation) teaches the aforementioned.
Regarding claim 4, Dicker further teaches a computer-implemented method comprising:
A) wherein the amount of time of interaction metric is based on viewing or listening times (Paragraph 147).
The examiner notes that Dicker teaches “wherein the amount of time of interaction metric is based on viewing or listening times” as “In step 184, the process 52 weights each similar items list based on the duration since the associated popular item was purchased by the user (with recently-purchased items weighted more heavily), or if the popular item was not purchased, the rating given to the popular item by the user” (Paragraph 147). The examiner further notes that duration teaches the claimed viewing times in the broadest reasonable interpretation.
Regarding claim 5, Dicker further teaches a computer-implemented method comprising:
A) wherein the amount of time of interaction metric is based on website or advertisement viewing times (Paragraph 147).
The examiner notes that Dicker teaches “wherein the amount of time of interaction metric is based on website or advertisement viewing times” as “In step 184, the process 52 weights each similar items list based on the duration since the associated popular item was purchased by the user (with recently-purchased items weighted more heavily), or if the popular item was not purchased, the rating given to the popular item by the user” (Paragraph 147). The examiner further notes that duration of viewing a product (i.e. a website) teaches the claimed viewing times in the broadest reasonable interpretation.
Regarding claims 8 and 14, Dicker further teaches a computer-implemented method and computer-implemented system comprising:
A) wherein the similarity score is calculated based on a comparison of metrics of the particular item and metrics of each of the other items (Paragraphs 76 and 99, Figure 1).
The examiner notes that Dicker teaches “wherein the similarity score is calculated based on a comparison of metrics of the particular item and metrics of each of the other items” as “Each similar items list 64 consists of the N (e.g., 20) items which, based on correlations between purchases of items, are deemed to be the most closely related to the respective popular item 62. Each item in the similar items list 64 is stored together with a commonality index ("CI") value which indicates the relatedness of that item to the popular item 62, based on sales of the respective items. A relatively high commonality index for a pair of items ITEM A and ITEM B indicates that a relatively large percentage of users who bought ITEM A also bought ITEM B (and vice versa). A relatively low commonality index for ITEM A and ITEM B indicates that a relatively small percentage of the users who bought ITEM A also bought ITEM B (and vice versa). As described below, the similar items lists are generated, for each popular item, by selecting the N other items that have the highest commonality index values” (Paragraph 76) and “In step 110, the process generates the commonality indexes for each (popular_item, other_item) pair in the table 108A. As indicated above, the commonality index (CI) values are measures of the similarity between two items, with larger CI values indicating greater degrees of similarity. The commonality indexes are preferably generated such that, for a given popular_item, the respective commonality indexes of the corresponding other_items take into consideration both (a) the number of customers that are common to both items, and (b) the total number of customers of the other_item. A preferred method for generating the commonality index values is set forth in equation (1) below, where N.sub.common is the number of users who purchased both A and B, sqrt is a square-root operation, N.sub.A is the number of users who purchased A, and N.sub.B is the number of users who purchased B” (Paragraph 99). The examiner further notes that computing commonality index values between each popular item 62 (i.e. the claimed particular item) and other items (see similar items list 64 in Figure 1) teaches the claimed similarity score as the use of the number of users who bought each item (i.e. metrics) are compared to arrive at a similarity score.
Regarding claims 10 and 16, Dicker further teaches a computer-implemented method and computer-implemented system comprising:
A) wherein the particular item is accessible as a video download, streaming video, a physical video copy, an audio download, streaming audio, a physical audio copy, an image, a game, a physical book, or an electronic book (Paragraph 42).
The examiner notes that Dicker teaches “wherein the particular item is accessible as a video download, streaming video, a physical video copy, an audio download, streaming audio, a physical audio copy, an image, a game, a physical book, or an electronic book” as “the merchant Web site includes functionality for allowing users to search, browse, and make purchases from an online catalog of purchasable items or "products," such as book titles, music titles, video titles, toys, and electronics products” (Paragraph 42). The examiner further notes that video titles, music titles, and book titles teach the claimed physical video copy, physical audio copy, and physical book respectively.
Regarding 15, Dicker further teaches a computer-implemented system comprising:
A) wherein the data output is a ranking of a particular person’s predicted preferences toward the other items (Paragraph 195, Figure 12).
The examiner notes that Dicker teaches “wherein the data output is a ranking of a particular person’s predicted preferences toward the other items” as “FIG. 13 illustrates a process that may be used to generate a related items list 500 of the type shown in FIG. 12. As illustrated, the related items list 500 for a given product is generated by retrieving the corresponding similar items list 64 (preferably from a viewing-history-based similar items table 60 as described above), optionally filtering out items falling outside the product category of the product, and then extracting the N top-rank items. Once this related items list 64 has been generated for a particular product, it may be re-used (e.g., cached) until the relevant similar items table 60 is regenerated” (Paragraph 195). The examiner further notes that outputting a ranked list of recommendations teaches the claimed predicted preferences.
Regarding claim 19, Dicker further teaches a computer-implemented method comprising:
A) wherein the output of the predictive model represents a particular person’s predicted preference toward the one of the other items (Paragraph 195, Figure 12).
The examiner notes that Dicker teaches “wherein the output of the predictive model represents a particular person’s predicted preference toward the one of the other items” as “FIG. 13 illustrates a process that may be used to generate a related items list 500 of the type shown in FIG. 12. As illustrated, the related items list 500 for a given product is generated by retrieving the corresponding similar items list 64 (preferably from a viewing-history-based similar items table 60 as described above), optionally filtering out items falling outside the product category of the product, and then extracting the N top-rank items. Once this related items list 64 has been generated for a particular product, it may be re-used (e.g., cached) until the relevant similar items table 60 is regenerated” (Paragraph 195). The examiner further notes that outputting a ranked list of recommendations teaches the claimed output.
Regarding claim 20, Dicker further teaches a computer-implemented method comprising:
A) wherein the particular person’s predicted preference toward the one of the other items is based on an amount of time of interaction metric with respect to the particular item, and the similarity score between the one or the other items and the particular item (Paragraphs 147, 195, Figure 12).
The examiner notes that Dicker teaches “wherein the particular person’s predicted preference toward the one of the other items is based on an amount of time of interaction metric with respect to the particular item, and the similarity score between the one or the other items and the particular item” as “In step 184, the process 52 weights each similar items list based on the duration since the associated popular item was purchased by the user (with recently-purchased items weighted more heavily), or if the popular item was not purchased, the rating given to the popular item by the user” (Paragraph 147), “FIG. 13 illustrates a process that may be used to generate a related items list 500 of the type shown in FIG. 12. As illustrated, the related items list 500 for a given product is generated by retrieving the corresponding similar items list 64 (preferably from a viewing-history-based similar items table 60 as described above), optionally filtering out items falling outside the product category of the product, and then extracting the N top-rank items. Once this related items list 64 has been generated for a particular product, it may be re-used (e.g., cached) until the relevant similar items table 60 is regenerated” (Paragraph 195). The examiner further notes that outputting a ranked list of recommendations (which is based on the items list 64 (which is based on calculated similarity scores (which can be based off of duration metrics (i.e. an amount of time of interaction metric)) teaches the aforementioned.
Regarding claims 21-23, Dicker further teaches a computer-implemented method, computer-implemented system, and non-transitory computer-readable medium comprising:
A) wherein the access history comprises interactions related to the particular item and the other items (Paragraph 74).
The examiner notes that Dicker teaches “wherein the particular item is accessible as a video download, streaming video, a physical video copy, an audio download, streaming audio, a physical audio copy, an image, a game, a physical book, or an electronic book” as “Each entry in the similar items table 60 is preferably in the form of a mapping of a popular item 62 to a corresponding list 64 of similar items ("similar items lists"). As used herein, a "popular" item is an item which satisfies some pre-specified popularity criteria. For example, in the embodiment described herein, an item is treated as popular of it has been purchased by more than 30 customers during the life of the Web site. Using this criteria produces a set of popular items (and thus a recommendation service) which grows over time. The similar items list 64 for a given popular item 62 may include other popular items” (Paragraph 74). The examiner further notes that a purchase history (i.e. an “access history”) of commercial items teaches the claimed interactions in the broadest reasonable interpretation.
Regarding claim 29, Dicker teaches a computer-implemented method comprising:
A) receiving data from a data storage device associated with a catalog of items and a particular item (Paragraph 58);
B) calculating, by digital electronic circuitry, similarity scores between the particular item and other items in the catalog (Paragraph 76, Figure 1);
C) retrieving, from data stores, an access history and associations associated with the particular item and the other items in the catalog (Paragraphs 58, 74, 129-135, and 209);
H) providing the access history and the associations associated with the particular item and the other items in the catalog to a predictive model (Paragraphs 74, 97, 99, and 128-135);
J) determining, by the digital electronic circuitry, weights of the predictive model based on the access histories and the associations for the particular item and the other items in view of the similarity scores (Paragraphs 85 and 128-135);
K) outputting data associated with one of the other items based on an output of the predictive model (Paragraphs 193 and 195, Figure 12); and
L) transmitting the one of the other items over a computer network for viewing by a user at a network access point (Paragraphs 193 and 195, Figure 12);
M) wherein the associations comprise artifacts collected by a content management system that relate to certain items to information about the user (Paragraphs 58 and 85).
The examiner notes that Dicker teaches “receiving data from a data storage device associated with a catalog of items and a particular item” as “The Web site 30 also includes a "user profiles" database 38 which stores account-specific information about users of the site. Because a group of individuals can share an account, a given "user" from the perspective of the Web site may include multiple actual users. As illustrated by FIG. 1, the data stored for each user may include one or more of the following types of information (among other things) that can be used to generate recommendations: (a) the user's purchase history, including dates of purchase, (b) a history of items recently viewed by the user, (c) the user's item ratings profile (if any), (d) the current contents of the user's personal shopping cart(s), and (e) a listing of items that were recently (e.g., within the last six months) removed from the shopping cart(s) without being purchased ("recent shopping cart contents")” (Paragraph 58). The examiner further notes that purchase history, viewing history, ratings, etc. teach the undefined claimed data. The examiner further notes that Dicker teaches “calculating, by digital electronic circuitry, similarity scores between the particular item and other items in the catalog” as “Each similar items list 64 consists of the N (e.g., 20) items which, based on correlations between purchases of items, are deemed to be the most closely related to the respective popular item 62. Each item in the similar items list 64 is stored together with a commonality index ("CI") value which indicates the relatedness of that item to the popular item 62, based on sales of the respective items. A relatively high commonality index for a pair of items ITEM A and ITEM B indicates that a relatively large percentage of users who bought ITEM A also bought ITEM B (and vice versa). A relatively low commonality index for ITEM A and ITEM B indicates that a relatively small percentage of the users who bought ITEM A also bought ITEM B (and vice versa). As described below, the similar items lists are generated, for each popular item, by selecting the N other items that have the highest commonality index values” (Paragraph 76). The examiner further notes that computing commonality index values between each popular item 62 (i.e. the claimed particular item) and other items (see similar items list 64 in Figure 1) teaches the claimed calculating. The examiner further notes that Dicker teaches “retrieving, from data stores, an access history and associations associated with the particular item and the other items in the catalog” as “The Web site 30 also includes a "user profiles" database 38 which stores account-specific information about users of the site. Because a group of individuals can share an account, a given "user" from the perspective of the Web site may include multiple actual users. As illustrated by FIG. 1, the data stored for each user may include one or more of the following types of information (among other things) that can be used to generate recommendations: (a) the user's purchase history, including dates of purchase, (b) a history of items recently viewed by the user, (c) the user's item ratings profile (if any), (d) the current contents of the user's personal shopping cart(s), and (e) a listing of items that were recently (e.g., within the last six months) removed from the shopping cart(s) without being purchased ("recent shopping cart contents")” (Paragraph 58), “Each entry in the similar items table 60 is preferably in the form of a mapping of a popular item 62 to a corresponding list 64 of similar items ("similar items lists"). As used herein, a "popular" item is an item which satisfies some pre-specified popularity criteria. For example, in the embodiment described herein, an item is treated as popular of it has been purchased by more than 30 customers during the life of the Web site. Using this criteria produces a set of popular items (and thus a recommendation service) which grows over time. The similar items list 64 for a given popular item 62 may include other popular items” (Paragraph 74), “The process flows shown in FIGS. 3A and 3B differ primarily in that they use different types of user actions as evidence of users' interests in a particular items. In the method shown in FIG. 3A, a user is assumed to be interested in an item if the user purchased the item; and in the process shown in 3B, a user is assumed to be interested in an item if the user viewed the item. Any of a variety of other types of user actions that evidence a user's interest in a particular item may additionally or alternatively be used, alone or in combination, to generate the similar items table 60. The following are examples of other types of user actions that may be used for this purpose:… (4) Submitting a favorable review for an item. With this method, products A and B may be treated as related if a large percentage of those favorably reviewed A also favorably reviewed B. A favorable review may be defined as a score that satisfies a particular threshold (e.g., 4 or above on a scale of 1-5)… With the above and other types of item-affinity-evidencing actions, equation (1) above may be used to generate the CI values” (Paragraphs 129-135), and “In step 650 the user's profile, or a portion of the profile, is read from a user database 38 (FIG. 1) or cache” (Paragraph 209). The examiner further notes that ascertaining popularity of the commercial items entails determining a purchase history (i.e. an “access history” in the broadest reasonable interpretation) of those commercial items. Moreover, retrieved user profile information includes various of amounts of data (such as ratings profile data (i.e. the claimed associations in the broadest reasonable interpretation (See also Page 6, lines 10-15 of the instant specification that defines the claimed associations as including liked items from a user))) that can be retrieved from multiple stores such as a database and/or cache. The examiner further notes that Dicker teaches “providing the access history and the associations associated with the particular item and the other items in the catalog to a predictive model” as “Each entry in the similar items table 60 is preferably in the form of a mapping of a popular item 62 to a corresponding list 64 of similar items ("similar items lists"). As used herein, a "popular" item is an item which satisfies some pre-specified popularity criteria. For example, in the embodiment described herein, an item is treated as popular of it has been purchased by more than 30 customers during the life of the Web site. Using this criteria produces a set of popular items (and thus a recommendation service) which grows over time. The similar items list 64 for a given popular item 62 may include other popular items” (Paragraph 74), “In step 106, the process identifies the items that constitute "popular" items. This may be accomplished, for example, by selecting from the item-to-customers table 104A those items that were purchased by more than a threshold number (e.g., 30) of customers. In the context of a merchant Web site such as that of Amazon.com, Inc., the resulting set of popular items may contain hundreds of thousands or millions of items” (Paragraph 97), “In step 110, the process generates the commonality indexes for each (popular_item, other_item) pair in the table 108A. As indicated above, the commonality index (CI) values are measures of the similarity between two items, with larger CI values indicating greater degrees of similarity” (Paragraph 99), “The process flows shown in FIGS. 3A and 3B differ primarily in that they use different types of user actions as evidence of users' interests in a particular items. In the method shown in FIG. 3A, a user is assumed to be interested in an item if the user purchased the item; and in the process shown in 3B, a user is assumed to be interested in an item if the user viewed the item. Any of a variety of other types of user actions that evidence a user's interest in a particular item may additionally or alternatively be used, alone or in combination, to generate the similar items table 60. The following are examples of other types of user actions that may be used for this purpose:… (4) Submitting a favorable review for an item. With this method, products A and B may be treated as related if a large percentage of those favorably reviewed A also favorably reviewed B. A favorable review may be defined as a score that satisfies a particular threshold (e.g., 4 or above on a scale of 1-5)… With the above and other types of item-affinity-evidencing actions, equation (1) above may be used to generate the CI values” (Paragraphs 129-135). The examiner further notes that providing purchase history data (i.e. access history) and/or review data (i.e. the claimed associations in the broadest reasonable interpretation) to the recommendation system of Dicker (i.e. a “predictive model”) results in the calculation of CI values for subsequent recommendations. The examiner further notes that Dicker teaches “determining, by the digital electronic circuitry, weights of the predictive model based on the access histories and the associations for the particular item and the other items in view of the similarity scores” as “In step 84, the similar items lists 64 are optionally weighted based on information about the user's affinity for the corresponding items of known interest. For example, a similar items list 64 may be weighted heavily if the user gave the corresponding popular item a rating of "5" on a scale or 1-5, or if the user purchased multiple copies of the item. Weighting a similar items list 64 heavily has the effect of increasing the likelihood that the items in that list we be included in the recommendations ultimately presented to the user. In one implementation described below, the user is presumed to have a greater affinity for recently purchased items over earlier purchased items. Similarly, where viewing histories are used to identify items of interest, items viewed recently may be weighted more heavily than earlier viewed items” (Paragraph 85) and “The process flows shown in FIGS. 3A and 3B differ primarily in that they use different types of user actions as evidence of users' interests in a particular items. In the method shown in FIG. 3A, a user is assumed to be interested in an item if the user purchased the item; and in the process shown in 3B, a user is assumed to be interested in an item if the user viewed the item. Any of a variety of other types of user actions that evidence a user's interest in a particular item may additionally or alternatively be used, alone or in combination, to generate the similar items table 60. The following are examples of other types of user actions that may be used for this purpose:… (4) Submitting a favorable review for an item. With this method, products A and B may be treated as related if a large percentage of those favorably reviewed A also favorably reviewed B. A favorable review may be defined as a score that satisfies a particular threshold (e.g., 4 or above on a scale of 1-5)… With the above and other types of item-affinity-evidencing actions, equation (1) above may be used to generate the CI values” (Paragraphs 129-135). The examiner further notes that a particular item and other items of a similar items list 64 are weighted via weights based off of user affinity data (that can include access histories and/or ratings data (i.e. an example of the claimed associations in the broadest reasonable interpretation)). The examiner further notes that Dicker teaches “outputting data associated with one of the other items based on an output of the predictive model” as “the table 60 may be used to display "canned" lists of related items on product detail pages of the "popular" items (i.e., items for which a similar items list 64 exists). FIG. 12 illustrates this feature in example form. In this example, the detail page of a product is supplemented with the message "customers who viewed this item also viewed the following items," followed by a hypertextual list 500 of four related items. In this particular embodiment, the list is generated from the viewing-history-based version of the similar items table (generated as described in section IV-B)” (Paragraph 193) and “FIG. 13 illustrates a process that may be used to generate a related items list 500 of the type shown in FIG. 12. As illustrated, the related items list 500 for a given product is generated by retrieving the corresponding similar items list 64 (preferably from a viewing-history-based similar items table 60 as described above), optionally filtering out items falling outside the product category of the product, and then extracting the N top-rank items” (Paragraph 195). The examiner further notes that the displayed recommended similar items (See Figure 12) of Dicker teaches the claimed output. The examiner further notes that Dicker teaches “transmitting the one of the other items over a computer network for viewing by a user at a network access point” as “the table 60 may be used to display "canned" lists of related items on product detail pages of the "popular" items (i.e., items for which a similar items list 64 exists). FIG. 12 illustrates this feature in example form. In this example, the detail page of a product is supplemented with the message "customers who viewed this item also viewed the following items," followed by a hypertextual list 500 of four related items. In this particular embodiment, the list is generated from the viewing-history-based version of the similar items table (generated as described in section IV-B)” (Paragraph 193) and “FIG. 13 illustrates a process that may be used to generate a related items list 500 of the type shown in FIG. 12. As illustrated, the related items list 500 for a given product is generated by retrieving the corresponding similar items list 64 (preferably from a viewing-history-based similar items table 60 as described above), optionally filtering out items falling outside the product category of the product, and then extracting the N top-rank items” (Paragraph 195). The examiner further notes that the displayed recommended similar items (See Figure 12) of Dicker (which shows an Amazon web page) teaches the claimed transmission over a network for viewing by a user via their computing device (i.e. a network access point). The examiner further notes that Dicker teaches “wherein the associations comprise artifacts collected by a content management system that relate to certain items to information about the user” as “The Web site 30 also includes a "user profiles" database 38 which stores account-specific information about users of the site. Because a group of individuals can share an account, a given "user" from the perspective of the Web site may include multiple actual users. As illustrated by FIG. 1, the data stored for each user may include one or more of the following types of information (among other things) that can be used to generate recommendations: (a) the user's purchase history, including dates of purchase, (b) a history of items recently viewed by the user, (c) the user's item ratings profile (if any), (d) the current contents of the user's personal shopping cart(s), and (e) a listing of items that were recently (e.g., within the last six months) removed from the shopping cart(s) without being purchased ("recent shopping cart contents")” (Paragraph 58) and “In step 84, the similar items lists 64 are optionally weighted based on information about the user's affinity for the corresponding items of known interest. For example, a similar items list 64 may be weighted heavily if the user gave the corresponding popular item a rating of "5" on a scale or 1-5, or if the user purchased multiple copies of the item” (Paragraph 85). The examiner further notes that retrieved user profile information includes various of amounts of data (such as ratings profile data (i.e. the claimed associations in the broadest reasonable interpretation (See also Page 6, lines 10-15 of the instant specification that defines the claimed associations as including liked items from a user))).
Dicker does not explicitly teach:
D) wherein the calculating comprises: identifying terms associated with the particular item and the other items;
E) generating a vector for the particular item and the other items;
F) wherein the vector represents a frequency score for each term associated with the particular item and the other items;
G) normalizing each vector.
Hassanzadeh, however, teaches “wherein the calculating comprises: identifying terms associated with the particular item and the other items” as “The tf-idf cosine similarity is a well established measure in the IR community which leverages the vector space model. This measure determines the closeness of the input strings r1 and r2 by first transforming the strings into unit vectors and then measuring the angle between their corresponding vectors. The cosine similarity with tf-idf weights is given by… where wr1 (t) and wr2 (t) are the normalized tf-idf weights for each common token in r1 and r2 respectively. The normalized tf-idf weight of token t in a given string record r is defined as follows… where tfr(t) is the term frequency of token t within string r and idf(t) is the inverse document frequency with respect to the entire relation R” (Section 4.1.3) and “We matched 38,064 movie titles in our database with 25,424 movie titles from DBpedia using the similarity predicates described above” (Section 4.3.2), “generating a vector for the particular item and the other items” as “The tf-idf cosine similarity is a well established measure in the IR community which leverages the vector space model. This measure determines the closeness of the input strings r1 and r2 by first transforming the strings into unit vectors and then measuring the angle between their corresponding vectors. The cosine similarity with tf-idf weights is given by… where wr1 (t) and wr2 (t) are the normalized tf-idf weights for each common token in r1 and r2 respectively. The normalized tf-idf weight of token t in a given string record r is defined as follows… where tfr(t) is the term frequency of token t within string r and idf(t) is the inverse document frequency with respect to the entire relation R” (Section 4.1.3) and “We matched 38,064 movie titles in our database with 25,424 movie titles from DBpedia using the similarity predicates described above” (Section 4.3.2), “wherein the vector represents a frequency score for each term associated with the particular item and the other items” as “The tf-idf cosine similarity is a well established measure in the IR community which leverages the vector space model. This measure determines the closeness of the input strings r1 and r2 by first transforming the strings into unit vectors and then measuring the angle between their corresponding vectors. The cosine similarity with tf-idf weights is given by… where wr1 (t) and wr2 (t) are the normalized tf-idf weights for each common token in r1 and r2 respectively. The normalized tf-idf weight of token t in a given string record r is defined as follows… where tfr(t) is the term frequency of token t within string r and idf(t) is the inverse document frequency with respect to the entire relation R” (Section 4.1.3) and “We matched 38,064 movie titles in our database with 25,424 movie titles from DBpedia using the similarity predicates described above” (Section 4.3.2), and “normalizing each vector” as “The tf-idf cosine similarity is a well established measure in the IR community which leverages the vector space model. This measure determines the closeness of the input strings r1 and r2 by first transforming the strings into unit vectors and then measuring the angle between their corresponding vectors. The cosine similarity with tf-idf weights is given by… where wr1 (t) and wr2 (t) are the normalized tf-idf weights for each common token in r1 and r2 respectively. The normalized tf-idf weight of token t in a given string record r is defined as follows… where tfr(t) is the term frequency of token t within string r and idf(t) is the inverse document frequency with respect to the entire relation R” (Section 4.1.3) and “We matched 38,064 movie titles in our database with 25,424 movie titles from DBpedia using the similarity predicates described above” (Section 4.3.2).
The examiner further notes that the secondary reference of Hassanzadeh teaches the concept of identifying terms of items for subsequent generation of vectors (which are then normalized) representing those items. Such vectors are representative of TF-IDF. The combination would result in using such a normalized vectors in the item similarity of Dicker.
It would have been obvious to one of ordinary skill in the art at the time the invention was made to combine the teachings of the cited references because teaching Hassanzadeh’s would have allowed Dicker’s to provide a method for high accuracy and/or scalable similarity algorithms, as noted by Hassanzadeh (Section 4.1).
Dicker and Hassanzadeh do not explicitly teach:
I) wherein the predictive model comprises a second-order rank of associations.
Chand, however, teaches “wherein the predictive model comprises a second-order rank of associations” as “A method is provided to aggregate a plurality of affinity lists to generate a single aggregated affinity list representing predicted affinities of a particular item, to other items, under a plurality of conditions” (Abstract), “The inventors have discovered that, by considering both affinity and lift values, a plurality of affinity lists (which, for example, might be used to generate item-based recommendations) can be advantageously aggregated, to generate a personalized recommendation” (Paragraph 16), and “The aggregated list is sort based on the aggregated inverse distances… this final ranked-aggregated list may serve as a basis for making recommendations to Ryan of other artists in which Ryan is likely to be interested” (Paragraph 38).
The examiner further notes that the secondary reference of Chand teaches the concept of a recommendation system (i.e. a predictive model) including a second-order rank of associations (which is interpreted as simply an aggregation of affinities (See Page 7, lines 21-23 of the instant specification)). The combination would result in the predictive recommendation model of Dicker to also use the undefined claimed second-order rank of associations.
It would have been obvious to one of ordinary skill in the art at the time the invention was made to combine the teachings of the cited references because teaching Chand’s would have allowed Dicker’s and Hassanzadeh’s to provide a method for providing personalized recommendations based off of multiple user characteristics, as noted by Chand (Paragraph 16).
11. Claims 3 and 9 are rejected under pre-AIA 35 U.S.C. 103(a) as being unpatentable over of Claims 7, 27, and 28 are rejected under pre-AIA 35 U.S.C. 103(a) as being unpatentable over of Dicker et al. (U.S. PGPUB 2010/0191619) in view of Hassanzadeh et al. (Article entitled “Linked Movie Data Base”, dated 20 April 2009), and further in view of Chand et al. (U.S. PGPUB 2007/0276826) as applied to claims 1-2, 4-5, 8, 10-12, 15-17, 19-23, and 27-29 above, and further in view of Neuneier et al. (U.S. PGPUB 2009/0204573).
12. Regarding claim 3, Dicker, Hassanzadeh, and Chand do not explicitly teach a computer-implemented comprising:
A) wherein the amount of time of interaction metric for a particular person is relative to other people in a pool of people.
Neuneier, however, teaches “wherein the amount of time of interaction metric for a particular person is relative to other people in a pool of people” as “the duration an average user is expected to stay on a web page may be calculated by taking all durations and the text length into consideration allowing a comparison of the individual duration of a single user compared with the average time of all users” (Paragraph 36).
The examiner further notes that the secondary reference of Neuneier teaches that duration of a user can be compared to the durations of other users.
It would have been obvious to one of ordinary skill in the art at the time the invention was made to combine the teachings of the cited references because teaching Neuneier’s would have allowed Dicker’s, Hassanzadeh’s, and Chand’s to provide a method for making durations better comparable to one another, as noted by Neuneier (Paragraph 36).
Regarding claim 9, Dicker further teaches a computer-implemented method comprising:
A) wherein the data output is a ranking of the particular person’s predicted preferences toward the other items (Paragraph 195, Figure 12).
The examiner notes that Dicker teaches “wherein the data output is a ranking of the particular person’s predicted preferences toward the other items” as “FIG. 13 illustrates a process that may be used to generate a related items list 500 of the type shown in FIG. 12. As illustrated, the related items list 500 for a given product is generated by retrieving the corresponding similar items list 64 (preferably from a viewing-history-based similar items table 60 as described above), optionally filtering out items falling outside the product category of the product, and then extracting the N top-rank items. Once this related items list 64 has been generated for a particular product, it may be re-used (e.g., cached) until the relevant similar items table 60 is regenerated” (Paragraph 195). The examiner further notes that outputting a ranked list of recommendations teaches the claimed predicted preferences.
13. Claims 7, 27, and 28 are rejected under pre-AIA 35 U.S.C. 103(a) as being unpatentable over of Dicker et al. (U.S. PGPUB 2010/0191619) in view of Hassanzadeh et al. (Article entitled “Linked Movie Data Base”, dated 20 April 2009), and further in view of Chand et al. (U.S. PGPUB 2007/0276826) as applied to claims 1-2, 4-5, 8, 10-12, 15-17, 19-23, and 27-29 above, and further in view of Cunningham et al. (U.S. PGPUB 2009/0094159), and further in view of Cheng et al. (Article entitled “Statistics and Social Network of YouTube Videos”, dated 2008).
14. Regarding claims 7, 27, and 28, Dicker further teaches a computer-implemented method, computer-implemented system and non-transitory computer-readable medium comprising:
A) wherein an interaction comprises purchases or rentals of items, monetary transaction histories, bookmarks, ratings, reviews, tags (Paragraph 58).
The examiner notes that Dicker teaches “wherein an interaction comprises purchases or rentals of items, monetary transaction histories, bookmarks, ratings, reviews, tags” as “The Web site 30 also includes a "user profiles" database 38 which stores account-specific information about users of the site. Because a group of individuals can share an account, a given "user" from the perspective of the Web site may include multiple actual users. As illustrated by FIG. 1, the data stored for each user may include one or more of the following types of information (among other things) that can be used to generate recommendations: (a) the user's purchase history, including dates of purchase, (b) a history of items recently viewed by the user, (c) the user's item ratings profile (if any), (d) the current contents of the user's personal shopping cart(s), and (e) a listing of items that were recently (e.g., within the last six months) removed from the shopping cart(s) without being purchased ("recent shopping cart contents")” (Paragraph 58). The examiner further notes that stored purchase history data (i.e. the claimed purchases or rentals of items and monetary transaction histories), current contents of a shopping cart (i.e. the claimed bookmarks (which are undefined in the instant specification) and tags (which are undefined in the instant specification) in the broadest reasonable interpretation), and ratings data (i.e. the claimed reviews and ratings) teach the claimed interaction.
Dicker, Hassanzadeh, and Chand do not explicitly teach:
A) wherein the interaction comprises number of times watched.
Cunningham, however, teaches “wherein the interaction comprises number of times watched” as “The advertising system 600 includes a video storage component 601 that receives an assembled video 660 from client contributors 654. The client contributors 654 may provide both first (e.g., stock) videos and assembled videos. Stock videos may be advertisements provided by companies that wish to display the advertisements to viewers such as a client viewer 658. Information about the client contributors 654 is stored in a Contributor Accounts table 632. The table 632 may be a file stored in persistent storage, or a relational database, or data stored in any other format on a computer-readable medium. The Contributor Accounts table tracks the following information for each user who has contributed media content to the advertising system: the number of direct clicks (i.e., clicks on the user's portion(s) of a video while the portion is displayed), direct views (i.e., number of times the user's portion of a video has been viewed), indirect clicks (i.e., number of clicks on other users' portions of a video, where the user does not hold rights to the portion, but does hold rights to some other portion of the video), indirect views (i.e., views of other user's portions of a video to which the user holds rights), and an account balance, which is the total amount of currency that has been deposited in the user's account (since some initial time) as a result of use or viewing of portions of the video for which the user receives payments, minus the total amount withdrawn” (Paragraph 52).
The examiner further notes that although Dicker teaches the storage of various metrics, there is no explicit teaching of the number of times watched. Nevertheless, the secondary reference of Cunningham teaches the storage of a number of views (i.e. number of times watched) of a video. The combination would result in expanding the storage of type of metrics in Dicker.
It would have been obvious to one of ordinary skill in the art at the time the invention was made to combine the teachings of the cited references because teaching Cunningham’s would have allowed Dicker’s, Hassanzadeh’s, and Chand’s to provide a method for tracking video-related data, as noted by Cunningham (Paragraph 5).
Dicker, Hassanzadeh, Chand, and Cunningham do not explicitly teach:
A) wherein the interaction comprises number of references to the item.
Cheng, however, teaches “wherein the interaction comprises number of references to the item” as “Each video contains a series of metadata: video ID, uploader, date when it was added, category, length, user rating, number of views, ratings and comments, and a list of “related videos”” (Page 230) and “Number of Ratings” (Table 1). The examiner further notes that the secondary reference of Cheng teaches the concept of storing a number of ratings (i.e. the claimed number of references to an item). The combination would result in expanding the storage of type of metrics in Dicker.
It would have been obvious to one of ordinary skill in the art at the time the invention was made to combine the teachings of the cited references because teaching Cheng’s would have allowed Dicker’s, Hassanzadeh’s, Chand’s, and Cunningham’s to provide a method for enhancing the service quality of videos, as noted by Cheng (Page 229).
15. Claims 30 and 33-34 are rejected under pre-AIA 35 U.S.C. 103(a) as being unpatentable over of Dicker et al. (U.S. PGPUB 2010/0191619) in view of Hassanzadeh et al. (Article entitled “Linked Movie Data Base”, dated 20 April 2009), and further in view of Chand et al. (U.S. PGPUB 2007/0276826), and further in view of Gross (U.S. PGPUB 2004/0267604).
16. Regarding claim 30, Dicker teaches a computer-implemented method comprising:
A) receiving data from a data storage device associated with a catalog of items and a particular item (Paragraph 58);
B) calculating, by digital electronic circuitry, similarity scores between the particular item and other items in the catalog (Paragraph 76, Figure 1);
G) retrieving, from data stores, an access history and associations associated with the particular item and the other items in the catalog (Paragraphs 58, 74, 129-135, and 209);
H) providing the access history and the associations associated with the particular item and the other items in the catalog to a predictive model (Paragraphs 74, 97, 99, and 128-135);
J) determining, by the digital electronic circuitry, weights of the predictive model based on the access histories and the associations for the particular item and the other items in view of the similarity scores (Paragraphs 85 and 128-135);
L) wherein the associations comprise artifacts collected by a content management system that relate to certain items to information about the user (Paragraphs 58 and 85).
The examiner notes that Dicker teaches “receiving data from a data storage device associated with a catalog of items and a particular item” as “The Web site 30 also includes a "user profiles" database 38 which stores account-specific information about users of the site. Because a group of individuals can share an account, a given "user" from the perspective of the Web site may include multiple actual users. As illustrated by FIG. 1, the data stored for each user may include one or more of the following types of information (among other things) that can be used to generate recommendations: (a) the user's purchase history, including dates of purchase, (b) a history of items recently viewed by the user, (c) the user's item ratings profile (if any), (d) the current contents of the user's personal shopping cart(s), and (e) a listing of items that were recently (e.g., within the last six months) removed from the shopping cart(s) without being purchased ("recent shopping cart contents")” (Paragraph 58). The examiner further notes that purchase history, viewing history, ratings, etc. teach the undefined claimed data. The examiner further notes that Dicker teaches “calculating, by digital electronic circuitry, similarity scores between the particular item and other items in the catalog” as “Each similar items list 64 consists of the N (e.g., 20) items which, based on correlations between purchases of items, are deemed to be the most closely related to the respective popular item 62. Each item in the similar items list 64 is stored together with a commonality index ("CI") value which indicates the relatedness of that item to the popular item 62, based on sales of the respective items. A relatively high commonality index for a pair of items ITEM A and ITEM B indicates that a relatively large percentage of users who bought ITEM A also bought ITEM B (and vice versa). A relatively low commonality index for ITEM A and ITEM B indicates that a relatively small percentage of the users who bought ITEM A also bought ITEM B (and vice versa). As described below, the similar items lists are generated, for each popular item, by selecting the N other items that have the highest commonality index values” (Paragraph 76). The examiner further notes that computing commonality index values between each popular item 62 (i.e. the claimed particular item) and other items (see similar items list 64 in Figure 1) teaches the claimed calculating. The examiner further notes that Dicker teaches “retrieving, from data stores, an access history and associations associated with the particular item and the other items in the catalog” as “The Web site 30 also includes a "user profiles" database 38 which stores account-specific information about users of the site. Because a group of individuals can share an account, a given "user" from the perspective of the Web site may include multiple actual users. As illustrated by FIG. 1, the data stored for each user may include one or more of the following types of information (among other things) that can be used to generate recommendations: (a) the user's purchase history, including dates of purchase, (b) a history of items recently viewed by the user, (c) the user's item ratings profile (if any), (d) the current contents of the user's personal shopping cart(s), and (e) a listing of items that were recently (e.g., within the last six months) removed from the shopping cart(s) without being purchased ("recent shopping cart contents")” (Paragraph 58), “Each entry in the similar items table 60 is preferably in the form of a mapping of a popular item 62 to a corresponding list 64 of similar items ("similar items lists"). As used herein, a "popular" item is an item which satisfies some pre-specified popularity criteria. For example, in the embodiment described herein, an item is treated as popular of it has been purchased by more than 30 customers during the life of the Web site. Using this criteria produces a set of popular items (and thus a recommendation service) which grows over time. The similar items list 64 for a given popular item 62 may include other popular items” (Paragraph 74), “The process flows shown in FIGS. 3A and 3B differ primarily in that they use different types of user actions as evidence of users' interests in a particular items. In the method shown in FIG. 3A, a user is assumed to be interested in an item if the user purchased the item; and in the process shown in 3B, a user is assumed to be interested in an item if the user viewed the item. Any of a variety of other types of user actions that evidence a user's interest in a particular item may additionally or alternatively be used, alone or in combination, to generate the similar items table 60. The following are examples of other types of user actions that may be used for this purpose:… (4) Submitting a favorable review for an item. With this method, products A and B may be treated as related if a large percentage of those favorably reviewed A also favorably reviewed B. A favorable review may be defined as a score that satisfies a particular threshold (e.g., 4 or above on a scale of 1-5)… With the above and other types of item-affinity-evidencing actions, equation (1) above may be used to generate the CI values” (Paragraphs 129-135), and “In step 650 the user's profile, or a portion of the profile, is read from a user database 38 (FIG. 1) or cache” (Paragraph 209). The examiner further notes that ascertaining popularity of the commercial items entails determining a purchase history (i.e. an “access history” in the broadest reasonable interpretation) of those commercial items. Moreover, retrieved user profile information includes various of amounts of data (such as ratings profile data (i.e. the claimed associations in the broadest reasonable interpretation (See also Page 6, lines 10-15 of the instant specification that defines the claimed associations as including liked items from a user))) that can be retrieved from multiple stores such as a database and/or cache. The examiner further notes that Dicker teaches “providing the access history and the associations associated with the particular item and the other items in the catalog to a predictive model” as “Each entry in the similar items table 60 is preferably in the form of a mapping of a popular item 62 to a corresponding list 64 of similar items ("similar items lists"). As used herein, a "popular" item is an item which satisfies some pre-specified popularity criteria. For example, in the embodiment described herein, an item is treated as popular of it has been purchased by more than 30 customers during the life of the Web site. Using this criteria produces a set of popular items (and thus a recommendation service) which grows over time. The similar items list 64 for a given popular item 62 may include other popular items” (Paragraph 74), “In step 106, the process identifies the items that constitute "popular" items. This may be accomplished, for example, by selecting from the item-to-customers table 104A those items that were purchased by more than a threshold number (e.g., 30) of customers. In the context of a merchant Web site such as that of Amazon.com, Inc., the resulting set of popular items may contain hundreds of thousands or millions of items” (Paragraph 97), “In step 110, the process generates the commonality indexes for each (popular_item, other_item) pair in the table 108A. As indicated above, the commonality index (CI) values are measures of the similarity between two items, with larger CI values indicating greater degrees of similarity” (Paragraph 99), “The process flows shown in FIGS. 3A and 3B differ primarily in that they use different types of user actions as evidence of users' interests in a particular items. In the method shown in FIG. 3A, a user is assumed to be interested in an item if the user purchased the item; and in the process shown in 3B, a user is assumed to be interested in an item if the user viewed the item. Any of a variety of other types of user actions that evidence a user's interest in a particular item may additionally or alternatively be used, alone or in combination, to generate the similar items table 60. The following are examples of other types of user actions that may be used for this purpose:… (4) Submitting a favorable review for an item. With this method, products A and B may be treated as related if a large percentage of those favorably reviewed A also favorably reviewed B. A favorable review may be defined as a score that satisfies a particular threshold (e.g., 4 or above on a scale of 1-5)… With the above and other types of item-affinity-evidencing actions, equation (1) above may be used to generate the CI values” (Paragraphs 129-135). The examiner further notes that providing purchase history data (i.e. access history) and/or review data (i.e. the claimed associations in the broadest reasonable interpretation) to the recommendation system of Dicker (i.e. a “predictive model”) results in the calculation of CI values for subsequent recommendations. The examiner further notes that Dicker teaches “determining, by the digital electronic circuitry, weights of the predictive model based on the access histories and the associations for the particular item and the other items in view of the similarity scores” as “In step 84, the similar items lists 64 are optionally weighted based on information about the user's affinity for the corresponding items of known interest. For example, a similar items list 64 may be weighted heavily if the user gave the corresponding popular item a rating of "5" on a scale or 1-5, or if the user purchased multiple copies of the item. Weighting a similar items list 64 heavily has the effect of increasing the likelihood that the items in that list we be included in the recommendations ultimately presented to the user. In one implementation described below, the user is presumed to have a greater affinity for recently purchased items over earlier purchased items. Similarly, where viewing histories are used to identify items of interest, items viewed recently may be weighted more heavily than earlier viewed items” (Paragraph 85) and “The process flows shown in FIGS. 3A and 3B differ primarily in that they use different types of user actions as evidence of users' interests in a particular items. In the method shown in FIG. 3A, a user is assumed to be interested in an item if the user purchased the item; and in the process shown in 3B, a user is assumed to be interested in an item if the user viewed the item. Any of a variety of other types of user actions that evidence a user's interest in a particular item may additionally or alternatively be used, alone or in combination, to generate the similar items table 60. The following are examples of other types of user actions that may be used for this purpose:… (4) Submitting a favorable review for an item. With this method, products A and B may be treated as related if a large percentage of those favorably reviewed A also favorably reviewed B. A favorable review may be defined as a score that satisfies a particular threshold (e.g., 4 or above on a scale of 1-5)… With the above and other types of item-affinity-evidencing actions, equation (1) above may be used to generate the CI values” (Paragraphs 129-135). The examiner further notes that a particular item and other items of a similar items list 64 are weighted via weights based off of user affinity data (that can include access histories and/or ratings data (i.e. an example of the claimed associations in the broadest reasonable interpretation)). The examiner further notes that Dicker teaches “wherein the associations comprise artifacts collected by a content management system that relate to certain items to information about the user” as “The Web site 30 also includes a "user profiles" database 38 which stores account-specific information about users of the site. Because a group of individuals can share an account, a given "user" from the perspective of the Web site may include multiple actual users. As illustrated by FIG. 1, the data stored for each user may include one or more of the following types of information (among other things) that can be used to generate recommendations: (a) the user's purchase history, including dates of purchase, (b) a history of items recently viewed by the user, (c) the user's item ratings profile (if any), (d) the current contents of the user's personal shopping cart(s), and (e) a listing of items that were recently (e.g., within the last six months) removed from the shopping cart(s) without being purchased ("recent shopping cart contents")” (Paragraph 58) and “In step 84, the similar items lists 64 are optionally weighted based on information about the user's affinity for the corresponding items of known interest. For example, a similar items list 64 may be weighted heavily if the user gave the corresponding popular item a rating of "5" on a scale or 1-5, or if the user purchased multiple copies of the item” (Paragraph 85). The examiner further notes that retrieved user profile information includes various of amounts of data (such as ratings profile data (i.e. the claimed associations in the broadest reasonable interpretation (See also Page 6, lines 10-15 of the instant specification that defines the claimed associations as including liked items from a user))).
Dicker does not explicitly teach:
C) wherein the calculating comprises: identifying terms associated with the particular item and the other items;
D) generating a vector for the particular item and the other items;
E) wherein the vector represents a frequency score for each term associated with the particular item and the other items;
F) normalizing each vector.
Hassanzadeh, however, teaches “wherein the calculating comprises: identifying terms associated with the particular item and the other items” as “The tf-idf cosine similarity is a well established measure in the IR community which leverages the vector space model. This measure determines the closeness of the input strings r1 and r2 by first transforming the strings into unit vectors and then measuring the angle between their corresponding vectors. The cosine similarity with tf-idf weights is given by… where wr1 (t) and wr2 (t) are the normalized tf-idf weights for each common token in r1 and r2 respectively. The normalized tf-idf weight of token t in a given string record r is defined as follows… where tfr(t) is the term frequency of token t within string r and idf(t) is the inverse document frequency with respect to the entire relation R” (Section 4.1.3) and “We matched 38,064 movie titles in our database with 25,424 movie titles from DBpedia using the similarity predicates described above” (Section 4.3.2), “generating a vector for the particular item and the other items” as “The tf-idf cosine similarity is a well established measure in the IR community which leverages the vector space model. This measure determines the closeness of the input strings r1 and r2 by first transforming the strings into unit vectors and then measuring the angle between their corresponding vectors. The cosine similarity with tf-idf weights is given by… where wr1 (t) and wr2 (t) are the normalized tf-idf weights for each common token in r1 and r2 respectively. The normalized tf-idf weight of token t in a given string record r is defined as follows… where tfr(t) is the term frequency of token t within string r and idf(t) is the inverse document frequency with respect to the entire relation R” (Section 4.1.3) and “We matched 38,064 movie titles in our database with 25,424 movie titles from DBpedia using the similarity predicates described above” (Section 4.3.2), “wherein the vector represents a frequency score for each term associated with the particular item and the other items” as “The tf-idf cosine similarity is a well established measure in the IR community which leverages the vector space model. This measure determines the closeness of the input strings r1 and r2 by first transforming the strings into unit vectors and then measuring the angle between their corresponding vectors. The cosine similarity with tf-idf weights is given by… where wr1 (t) and wr2 (t) are the normalized tf-idf weights for each common token in r1 and r2 respectively. The normalized tf-idf weight of token t in a given string record r is defined as follows… where tfr(t) is the term frequency of token t within string r and idf(t) is the inverse document frequency with respect to the entire relation R” (Section 4.1.3) and “We matched 38,064 movie titles in our database with 25,424 movie titles from DBpedia using the similarity predicates described above” (Section 4.3.2), and “normalizing each vector” as “The tf-idf cosine similarity is a well established measure in the IR community which leverages the vector space model. This measure determines the closeness of the input strings r1 and r2 by first transforming the strings into unit vectors and then measuring the angle between their corresponding vectors. The cosine similarity with tf-idf weights is given by… where wr1 (t) and wr2 (t) are the normalized tf-idf weights for each common token in r1 and r2 respectively. The normalized tf-idf weight of token t in a given string record r is defined as follows… where tfr(t) is the term frequency of token t within string r and idf(t) is the inverse document frequency with respect to the entire relation R” (Section 4.1.3) and “We matched 38,064 movie titles in our database with 25,424 movie titles from DBpedia using the similarity predicates described above” (Section 4.3.2).
The examiner further notes that the secondary reference of Hassanzadeh teaches the concept of identifying terms of items for subsequent generation of vectors (which are then normalized) representing those items. Such vectors are representative of TF-IDF. The combination would result in using such a normalized vectors in the item similarity of Dicker.
Dicker and Hassanzadeh do not explicitly teach:
I) wherein the predictive model comprises a second-order rank of associations.
Chand, however, teaches “wherein the predictive model comprises a second-order rank of associations” as “A method is provided to aggregate a plurality of affinity lists to generate a single aggregated affinity list representing predicted affinities of a particular item, to other items, under a plurality of conditions” (Abstract), “The inventors have discovered that, by considering both affinity and lift values, a plurality of affinity lists (which, for example, might be used to generate item-based recommendations) can be advantageously aggregated, to generate a personalized recommendation” (Paragraph 16), and “The aggregated list is sort based on the aggregated inverse distances… this final ranked-aggregated list may serve as a basis for making recommendations to Ryan of other artists in which Ryan is likely to be interested” (Paragraph 38).
The examiner further notes that the secondary reference of Chand teaches the concept of a recommendation system (i.e. a predictive model) including a second-order rank of associations (which is interpreted as simply an aggregation of affinities (See Page 7, lines 21-23 of the instant specification)). The combination would result in the predictive recommendation model of Dicker to also use the undefined claimed second-order rank of associations.
It would have been obvious to one of ordinary skill in the art at the time the invention was made to combine the teachings of the cited references because teaching Chand’s would have allowed Dicker’s and Hassanzadeh’s to provide a method for providing personalized recommendations based off of multiple user characteristics, as noted by Chand (Paragraph 16).
Dicker, Hassanzadeh, and Chand do not explicitly teach:
K) identifying a particular person from a pool for the particular item based on an output of the predictive model indicating that the person was someone who interacted early with a similar, popular item.
Gross, however, teaches “identifying a particular person from a pool for the particular item based on an output of the predictive model indicating that the person was someone who interacted early with a similar, popular item” as “Yet another aspect concerns a method of presenting advertising to an online community, comprising the steps of: processing member historical records of content reviewed and/or items purchased for each member of the online community, and comparing such member historical records with other member historical records; identifying a first member as having a trendsetter status when the results indicate that such first member exhibits behavior which is imitated by other members; providing a recommendation to the first member within a first screen during an online session using a recommender system; and adjusting advertising presented to the member in the first screen based on whether the first member has a trendsettet status” (Paragraph 48), “FIG. 2A is a flow chart illustrating the steps performed by a trendsetter identification process implemented in accordance with one exemplary embodiment of the present invention. As seen there, a first step 210 examines which items are the most popular within the community at a given time, which may be the present, or some prior date. It should be apparent that the process can be executed to identify trendsetters for a single items, multiple items, or items within a larger logical grouping, such as a category or sub-category of items. For example, an item might be a particular title of a book; a category of books might be logically grouped by artist, genre, publisher, etc. It should be clear that "popularity" of an item (or items) could be measured by reference to numbers of units sold, a number of units rented, a number of page views, a number of queries, a number of messages, etc., and the degree by which an item is deemed to be popular can be measure in any number of ways, including, for example, a percentage. Thus, in the present example, an item is deemed "popular" when it is among the top 10, or among the top 10% of items. Other applications are likely to use other benchmarks for determining popularity” (Paragraphs 101-102) and “Finally, at step 240 the trendsetters are explicitly listed by item, by a group of items, or in aggregate across an entire sampling population. These lists can be used as noted below for private use in marketing, planning, and/or they can be published electronically online as well for community consumption. In the latter case a particular community can see who the trendsetters are for a particular item, or who the trendsetters are for a category of items, or who are the overall trendsetters across all items” (Paragraph 107).
The examiner further notes that the secondary reference of Gross teaches the concept of recommending items to ascertained trendsetters (i.e. the claimed particular person that is an early adopter).
It would have been obvious to one of ordinary skill in the art at the time the invention was made to combine the teachings of the cited references because teaching Gross’s would have allowed Dicker’s, Hassanzadeh’s, and Chand’s to provide a method for allowing users of an online community to benefit from the prescience of trendsetters, as noted by Gross (Paragraph 80).
Regarding claim 33, Dicker further teaches a computer-implemented method comprising:
A) wherein the weights of the predictive model are further adjusted based on an amount of time interaction metric (Paragraph 147).
The examiner notes that Dicker teaches “wherein the weights of the predictive model are further adjusted based on an amount of time interaction metric” as “In step 184, the process 52 weights each similar items list based on the duration since the associated popular item was purchased by the user (with recently-purchased items weighted more heavily), or if the popular item was not purchased, the rating given to the popular item by the user” (Paragraph 147). The examiner further notes that as explained in the 112 rejection above, the claimed adjusting of the weights is deemed new matter as there is no explicit support whatsoever for weights being adjusted by an amount of time interaction metric. Nevertheless, the weights of Dicker being based on a duration metric (i.e. the claimed amount of time interaction metric in the broadest reasonable interpretation) teaches the aforementioned.
Regarding claim 34, Dicker further teaches a computer-implemented method comprising:
A) wherein the similarity scores are calculated based on comparison of metrics of the particular item and metrics of each of the other items (Paragraphs 76 and 99, Figure 1).
The examiner notes that Dicker teaches “wherein the similarity scores are calculated based on comparison of metrics of the particular item and metrics of each of the other items” as “Each similar items list 64 consists of the N (e.g., 20) items which, based on correlations between purchases of items, are deemed to be the most closely related to the respective popular item 62. Each item in the similar items list 64 is stored together with a commonality index ("CI") value which indicates the relatedness of that item to the popular item 62, based on sales of the respective items. A relatively high commonality index for a pair of items ITEM A and ITEM B indicates that a relatively large percentage of users who bought ITEM A also bought ITEM B (and vice versa). A relatively low commonality index for ITEM A and ITEM B indicates that a relatively small percentage of the users who bought ITEM A also bought ITEM B (and vice versa). As described below, the similar items lists are generated, for each popular item, by selecting the N other items that have the highest commonality index values” (Paragraph 76) and “In step 110, the process generates the commonality indexes for each (popular_item, other_item) pair in the table 108A. As indicated above, the commonality index (CI) values are measures of the similarity between two items, with larger CI values indicating greater degrees of similarity. The commonality indexes are preferably generated such that, for a given popular_item, the respective commonality indexes of the corresponding other_items take into consideration both (a) the number of customers that are common to both items, and (b) the total number of customers of the other_item. A preferred method for generating the commonality index values is set forth in equation (1) below, where N.sub.common is the number of users who purchased both A and B, sqrt is a square-root operation, N.sub.A is the number of users who purchased A, and N.sub.B is the number of users who purchased B” (Paragraph 99). The examiner further notes that computing commonality index values between each popular item 62 (i.e. the claimed particular item) and other items (see similar items list 64 in Figure 1) teaches the claimed similarity score as the use of the number of users who bought each item (i.e. metrics) are compared to arrive at a similarity score.
Response to Arguments
17. Applicant’s arguments with respect to claims 1-5, 7-12, 14-17, 19-23, 27-30, and 33-34 have been considered but are moot because the new ground of rejection does not rely on any reference applied in the prior rejection of record for any teaching or matter specifically challenged in the argument (See newly applied reference of Chand).
Applicant's arguments filed 03/20/2026 have been fully considered but they are not persuasive.
Applicants argue on Page 12 that “These rejections conflict with the most recent guidance from the Patent Office. In a recent Appeals Review Panel decision, Director Squires stated, "Categorically excluding AI innovations from patent protection in the United States jeopardizes America's leadership in this critical emerging technology" in reaction to a panel decision which "essentially equated any machine learning with an unpatentable 'algorithm' and the remaining additional elements as "generic computer components,' without adequate explanation." Ex parte Desjardins, Appeal No. 2024-000567 (Sept. 26, 2025). Claim 1 is directed to a computer-implemented predictive model and falls within the class of inventions about which Director Squires has expressed concern”. However, unlike the claims in Desjardins (which were directed towards the improvement of the training of a machine learning model itself), the instant claims do not recite any sort machine-learning model whose training is improved. Rather, the claimed predictive model is interpreted as a generic algorithm that can be performed by a human via their mind and/or pen & paper. No actual machine-learning (or improvement to any training of such machine-learning) is reflected in the claims.
Applicants argue on Pages 13-14 that “Claim 1 also uses a combined order of specific rules (e.g., "calculating similarity scores"), which render information into a specific format (e.g., "generating a vector") that is then used and applied (e.g., "normalizing each vector" and "providing the associations to a predictive model") to create desired results (e.g., "outputting data predictive model")”. However, as explained in the abstract idea rejection above, calculation of similarity scores, generation of a vector, and providing associations to a predictive model (which is interpreted as a generic algorithm) can be performed by a human via their mind and/or pen & paper. Furthermore, the output of results is simply a mere data output operation that is an insignificant data outputting operation that does not integrate the abstract idea into a practical application.
Applicants argue on Page 14 that “Likewise, claim 1 takes a real-world, practical input (e.g., "access history and associations associated with the particular item and other items in the catalog"), processes that data, and transforms it into a real-world, practical output (e.g., "outputting data associated with one of the other items based on an output of the predictive model"). Thus, claim 1 is also patent eligible”. However, the instant claims are unlike in Diehr and are similar to claim 2 of example 47 that was deemed to be ineligible.
Applicants argue on Page 15 that “Just as a human mind is not equipped to detect suspicious activity by using network monitors, a human mind is not equipped to detect a current access of a particular item. Detecting suspicious activity is similar to detecting a current access of a particular item, because both involve detecting an activity on a network with high particularity”. However, unlike in SRI, the instant claims merely recite the detection of a current access of data that can be performed by a human via their mind and/or pen & paper.
Applicants argue on Page 15 that “Similarly, just as the human mind is not equipped to analyze network packets, it is also not equipped to generate vectors representing frequency scores, normalize vectors, or determine weights of a predictive model. This type of analysis can be performed with pen and paper, at least in theory. But analyzing network packets is also a task that could theoretically be performed with pen and paper. In fact, every algorithm can theoretically be performed via pen & paper”. However, a human can vectorize data, normalize such vectorized data, and determine weights via their mind and/or pen & paper.
Conclusion
18. The prior art made of record and not relied upon is considered pertinent to applicant's disclosure.
U.S. Patent 7,483,846 issued to Kumar et al. on 27 January 2009. The subject matter disclosed therein is pertinent to that of claims 1-5, 7-12, 14-17, 19-23, 27-30, and 33-34 (e.g., methods to identify potential influencers).
U.S. PGPUB 2011/0282758 issued to Jacobi et al. on 17 Novembers 2011. The subject matter disclosed therein is pertinent to that of claims 1-5, 7-12, 14-17, 19-23, 27-30, and 33-34 (e.g., methods to identify potential influencers).
Contact Information
19. Any inquiry concerning this communication or earlier communications from the examiner should be directed to Mahesh Dwivedi whose telephone number is (571) 272-2731. The examiner can normally be reached on Monday to Friday 8:20 am – 4:40 pm.
If attempts to reach the examiner by telephone are unsuccessful, the examiner’s supervisor, Charles Rones can be reached (571) 272-4085. The fax number for the organization where this application or proceeding is assigned is (571) 273-8300.
Information regarding the status of an application may be obtained from the Patent Application Information Retrieval (PAIR) system. Status information for published applications may be obtained from either Private PAIR or Public PAIR. Status information for unpublished applications is available through Private PAIR only. For more information about the PAIR system, see http://pair-direct.uspto.gov. Should you have questions on access to the Private PAIR system, contact the Electronic Business Center (EBC) at 866-217-9197 (toll-free).
Mahesh Dwivedi
Primary Examiner
Art Unit 2168
March 27, 2026
/MAHESH H DWIVEDI/Primary Examiner, Art Unit 2168
Prosecution Insights