DETAILED ACTION
Notice of Pre-AIA or AIA Status
The present application, filed on or after March 16, 2013, is being examined under the first inventor to file provisions of the AIA .
Information Disclosure Statement
The information disclosure statement (IDS) submitted on August 1, 2024, was considered by the examiner. The submission is in compliance with the provisions of 37 CFR 1.97.
Claim Rejections - 35 USC § 101
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.
Claims 1-21 are rejected under 35 U.S.C. 101 because the claimed invention is directed to an abstract idea (mental process) without significantly more.
Claim 1:
Regarding claim 1, in step 1 of the 101-analysis set forth in MPEP 2106, the claim recites
“A method of interacting with a customer on an e-commerce platform, the method comprising: … , and a method is one of the four statutory categories of invention.
In step 2A prong 1 of the 101-analysis set forth in the MPEP 2106, the examiner has determined that the following limitations recite a process that, under the broadest reasonable interpretation, covers a mental process but for recitation of generic computer components:
… rank a plurality of messages based on a set of metrics; (mental process, a person can mentally evaluate and rank messages based on metrics, see MPEP 2106.04(a)(2)(III)),
evaluating, … whether the customer would respond to one out of the plurality of messages, (mental process, a person can mentally evaluate and assess if a customer replies to messages, see MPEP 2106.04(a)(2)(III)),
determining, … , a resulting action from the customer; (mental process, a person can mentally evaluate and determine a customer’s resulting action, see MPEP 2106.04(a)(2)(III)),
If claim limitations, under their broadest reasonable interpretation, covers performance of the limitations as a mental process but for the recitation of generic computer components, then it falls within the mental process grouping of abstract ideas. Accordingly, the claim “recites” an abstract idea.
In step 2A prong 2 of the 101-analysis set forth in MPEP 2106, the examiner has determined that the following additional elements do not integrate this judicial exception into a practical application:
A method of interacting with a customer on an e-commerce platform, the method comprising: … providing, by a processor, a model … (In step 2A, prong 2, a processor recites a generic computer component being used as a tool. Providing a model recites mere instructions to apply an exception using generic computer – see MPEP 2106.05(f)),
evaluating, by the processor, … (In step 2A, prong 2, a processor recites a generic computer component being used as a tool – see MPEP 2106.05(f)),
recommending, by the processor, at least one out of the plurality of messages that results in the highest probability of response from the customer; (Mere instructions to apply an exception using generic computer – see MPEP 2106.05(f)),
sending, by the processor, the recommended one out of the plurality of messages to the customer; (In step 2A, prong 2, sending recites mere data transmission, which is considered insignificant extra-solution activity – see MPEP 2106.05(g)),
determining, by the processor, … (In step 2A, prong 2, a processor recites a generic computer component being used as a tool – see MPEP 2106.05(f)),
and updating, by the processor, the set of metrics based on the resulting action to further train the model, (Mere instructions to apply an exception using generic computer – see MPEP 2106.05(f)),
Since the claim as a whole, looking at the additional elements individually and in combination, does not contain any other additional elements that are indicative of integration into a practical application, the claim is “directed” to an abstract idea.
In step 2B of the 101-analysis set forth in the 2019 PEG, the examiner has determined that the claim does not include additional elements that are sufficient to amount to significantly more than the judicial exception.
As discussed above, additional elements iv, v, and viii recite a generic computer component being used as a tool, and additional elements vi and ix recite mere instructions to apply the judicial exception using generic computer components, which are not indicative of significantly more. The additional element vii recites mere data gathering and is considered as insignificant extra-solution activity. In step 2B, this insignificant extra-solution activity is well understood routine and conventional activity, which includes receiving or transmitting data over a network from court case Symantec, 838 F.3d at 1321, 120 USPQ2d at 1362 (utilizing an intermediary computer to forward information); TLI Communications LLC v. AV Auto. LLC, 823 F.3d 607, 610, 118 USPQ2d 1744, 1745 (Fed. Cir. 2016), – see MPEP 2106.05(d) (II)(i)).
Considering the additional elements individually and in combination, and the claim as a whole, the additional elements do not provide significantly more than the abstract idea. Therefore, the claim is not patent eligible.
Claim 2:
Regarding claim 2, it is dependent upon claim 1, and thereby incorporates the limitations of, and corresponding analysis applied to claim 1. Further, claim 2 recites the following additional element:
The method of claim 1, wherein said model is a machine learning model. (In step 2A, prong 2, this is considered mere instructions to apply an exception using generic computer – see MPEP 2106.05(f)). (In step 2B, this is also considered mere instructions to implement an abstract idea using generic computer – see MPEP 2106.05(f)).
Since the claim does not recite additional elements that either integrate the judicial exception into a practical application, nor provide significantly more than the judicial exception, the claim is not patent eligible.
Claim 3:
Regarding claim 3, it is dependent upon claim 2, and thereby incorporates the limitations of, and corresponding analysis applied to claim 2. Further, claim 3 recites the following additional element:
The method of claim 2, wherein the set of metrics contains information about the e-commerce platform, information about the customer; and information about the customer browsing session, (In step 2A, prong 2, this recites an indication to a field of use or technological environment – see MPEP 2106.05(h)), (In step 2B, this also recites a field of use or technological environment – see MPEP 2106.05(h)),
Since the claim does not recite additional elements that either integrate the judicial exception into a practical application, nor provide significantly more than the judicial exception, the claim is not patent eligible.
Claim 4:
Regarding claim 4, it is dependent upon claim 3, and thereby incorporates the limitations of, and corresponding analysis applied to claim 3. Further, claim 4 recites the following additional element:
The method of claim 3, wherein the training of the machine learning model is completed using at least one of: inverse propensity-scoring algorithm; doubly robust algorithm; and importance weighted regression algorithm, (In step 2A, prong 2, this is considered mere instructions to apply an exception using generic computer – see MPEP 2106.05(f)). (In step 2B, this is also considered mere instructions to apply an exception using generic computer – see MPEP 2106.05(f)).
Since the claim does not recite additional elements that either integrate the judicial exception into a practical application, nor provide significantly more than the judicial exception, the claim is not patent eligible.
Claim 5:
Regarding claim 5, it is dependent upon claim 4, and thereby incorporates the limitations of, and corresponding analysis applied to claim 4. Further, claim 5 recites the following abstract idea:
The method of claim 4, wherein the method further comprises the step of amending at least one of the plurality of messages if the message receives a low probability of response from the customer, (This recites a mental process, a person can mentally evaluate and amend or update messages if the message receives a low probability of response from the customer, see MPEP 2106.04(a)(2)(III)),
Since the claim does not recite additional elements that either integrate the judicial exception into a practical application, nor provide significantly more than the judicial exception, the claim is not patent eligible.
Claim 6:
Regarding claim 6, it is dependent upon claim 5, and thereby incorporates the limitations of, and corresponding analysis applied to claim 5. Further, claim 6 recites the following additional element:
The method of claim 5, wherein the plurality of messages comprises a text-based prompt; voice-based prompt; image based prompt; or video based prompt, (In step 2A, prong 2, this is considered mere instructions to apply an exception using generic computer – see MPEP 2106.05(f)). (In step 2B, this is also considered mere instructions to apply an exception using generic computer – see MPEP 2106.05(f)).
Since the claim does not recite additional elements that either integrate the judicial exception into a practical application, nor provide significantly more than the judicial exception, the claim is not patent eligible.
Claim 7:
Regarding claim 7, it is dependent upon claim 6, and thereby incorporates the limitations of, and corresponding analysis applied to claim 6. Further, claim 7 recites the following additional element:
The method of claim 6, wherein the set of metrics further comprises information about browsing behavior; recurrency; provenance; geolocation; and temporal details. (In step 2A, prong 2, this recites an indication to a field of use or technological environment – see MPEP 2106.05(h)), (In step 2B, this also recites a field of use or technological environment – see MPEP 2106.05(h)),
Since the claim does not recite additional elements that either integrate the judicial exception into a practical application, nor provide significantly more than the judicial exception, the claim is not patent eligible.
Claim 8:
Regarding claim 8, it is dependent upon claim 7, and thereby incorporates the limitations of, and corresponding analysis applied to claim 7. Further, claim 8 recites the following additional element:
The method of claim 7, wherein the resulting action from the customer is a positive action such that the customer accepts the message, (In step 2A, prong 2, this is considered mere instructions to apply an exception using generic computer – see MPEP 2106.05(f)). (In step 2B, this is also considered mere instructions to apply an exception using generic computer – see MPEP 2106.05(f)).
Since the claim does not recite additional elements that either integrate the judicial exception into a practical application, nor provide significantly more than the judicial exception, the claim is not patent eligible.
Claim 9:
Regarding claim 9, it is dependent upon claim 8, and thereby incorporates the limitations of, and corresponding analysis applied to claim 8. Further, claim 9 recites the following additional element:
The method of claim 8, wherein the resulting action from the customer is a negative action such that the customer declines the message, (In step 2A, prong 2, this is considered mere instructions to apply an exception using generic computer – see MPEP 2106.05(f)). (In step 2B, this is also considered mere instructions to apply an exception using generic computer – see MPEP 2106.05(f)).
Since the claim does not recite additional elements that either integrate the judicial exception into a practical application, nor provide significantly more than the judicial exception, the claim is not patent eligible.
Claim 10:
Regarding claim 10, it is dependent upon claim 9, and thereby incorporates the limitations of, and corresponding analysis applied to claim 9. Further, claim 10 recites the following additional element:
The method of claim 9, wherein the resulting action from the customer is a neutral action such that the customer ignores the message, (In step 2A, prong 2, this is considered mere instructions to apply an exception using generic computer – see MPEP 2106.05(f)). (In step 2B, this is also considered mere instructions to apply an exception using generic computer – see MPEP 2106.05(f)).
Since the claim does not recite additional elements that either integrate the judicial exception into a practical application, nor provide significantly more than the judicial exception, the claim is not patent eligible.
Claim 11:
Regarding claim 11, in step 1 of the 101-analysis set forth in MPEP 2106, the claim recites
“11. An apparatus, comprising: a memory for storing instructions; and a processor configured to execute the instructions and thereby cause the apparatus to at least: ...”, and an apparatus or a system is one of the four statutory categories of invention.
In step 2A prong 1 of the 101-analysis set forth in the MPEP 2106, the examiner has determined that the following limitations recite a process that, under the broadest reasonable interpretation, covers a mental process but for recitation of generic computer components:
… rank a plurality of messages based on a set of metrics; (mental process, a person can mentally evaluate and rank messages based on metrics, see MPEP 2106.04(a)(2)(III)),
evaluate and/or predict whether the customer would respond to one out of the plurality of messages, (mental process, a person can mentally evaluate and predict if a customer replies to messages, see MPEP 2106.04(a)(2)(III)),
determine a resulting action from the customer, (mental process, a person can mentally evaluate and determine a customer’s resulting action, see MPEP 2106.04(a)(2)(III)),
If claim limitations, under their broadest reasonable interpretation, covers performance of the limitations as a mental process but for the recitation of generic computer components, then it falls within the mental process grouping of abstract ideas. Accordingly, the claim “recites” an abstract idea.
In step 2A prong 2 of the 101-analysis set forth in MPEP 2106, the examiner has determined that the following additional elements do not integrate this judicial exception into a practical application:
An apparatus, comprising: a memory for storing instructions; and a processor configured to execute the instructions and thereby cause the apparatus to at least provide a model … (In step 2A, prong 2, a processor and memory recite generic computer components being used as a tool. Provide a model recites mere instructions to apply an exception using generic computer – see MPEP 2106.05(f)),
recommend at least one out of the plurality of messages that results in the highest probability of response from the customer; (Mere instructions to apply an exception using generic computer – see MPEP 2106.05(f)),
send, …, the recommended one out of the plurality of messages to the customer; (In step 2A, prong 2, sending recites mere data transmission, which is considered insignificant extra-solution activity – see MPEP 2106.05(g)),
… via an interface … said interface being adapted to display the message; (Mere instructions to apply an exception using generic computer – see MPEP 2106.05(f)),
and update the set of metrics based on the resulting action to further train the model, (Mere instructions to apply an exception using generic computer – see MPEP 2106.05(f)),
Since the claim as a whole, looking at the additional elements individually and in combination, does not contain any other additional elements that are indicative of integration into a practical application, the claim is “directed” to an abstract idea.
In step 2B of the 101-analysis set forth in the 2019 PEG, the examiner has determined that the claim does not include additional elements that are sufficient to amount to significantly more than the judicial exception.
As discussed above, additional element iv recites generic computer components being used as a tool. Additional elements v, vii, and viii recite mere instructions to apply the judicial exception using generic computer components, which are not indicative of significantly more. The additional element vi recites mere data gathering or transmission, and is considered insignificant extra-solution activity. In step 2B, this insignificant extra-solution activity is well understood routine and conventional activity, which includes receiving or transmitting data over a network from court case Symantec, 838 F.3d at 1321, 120 USPQ2d at 1362 (utilizing an intermediary computer to forward information); TLI Communications LLC v. AV Auto. LLC, 823 F.3d 607, 610, 118 USPQ2d 1744, 1745 (Fed. Cir. 2016), – see MPEP 2106.05(d) (II)(i)),
Considering the additional elements individually and in combination, and the claim as a whole, the additional elements do not provide significantly more than the abstract idea. Therefore, the claim is not patent eligible.
Claims 12-20:
Claims 12 -20 recite similar limitations as corresponding claims 2-10 listed above, and are rejected for similar reasons under 35 U.S.C. 101.
Claim 21:
Regarding claim 21, in step 1 of the 101-analysis set forth in MPEP 2106, the claim recites
“A method of interacting with a customer on an e-commerce platform, the method comprising: … , and a method is one of the four statutory categories of invention.
In step 2A prong 1 of the 101-analysis set forth in the MPEP 2106, the examiner has determined that the following limitations recite a process that, under the broadest reasonable interpretation, covers a mental process but for recitation of generic computer components:
… generating … a plurality of messages based on a set of metrics associated with the customer; (mental process, a person can mentally evaluate and rank messages based on metrics, see MPEP 2106.04(a)(2)(III)),
evaluating, … whether the customer would respond to one out of the plurality of messages, (mental process, a person can mentally evaluate and assess if a customer replies to messages, see MPEP 2106.04(a)(2)(III)),
evaluating, … , a resulting action from the customer; (mental process, a person can mentally evaluate a customer’s resulting action, see MPEP 2106.04(a)(2)(III)),
If claim limitations, under their broadest reasonable interpretation, covers performance of the limitations as a mental process but for the recitation of generic computer components, then it falls within the mental process grouping of abstract ideas. Accordingly, the claim “recites” an abstract idea.
In step 2A prong 2 of the 101-analysis set forth in MPEP 2106, the examiner has determined that the following additional elements do not integrate this judicial exception into a practical application:
A method of interacting with a customer on an e-commerce platform, the method comprising: … by a processor, … (In step 2A, prong 2, a processor recites a generic computer component being used as a tool – see MPEP 2106.05(f)),
evaluating, by the processor, … (In step 2A, prong 2, a processor recites a generic computer component being used as a tool – see MPEP 2106.05(f)),
recommending, by the processor, at least one out of the plurality of messages that results in the highest probability of response from the customer; (Mere instructions to apply an exception using generic computer – see MPEP 2106.05(f)),
sending, by the processor, the recommended one out of the plurality of messages to the customer; (In step 2A, prong 2, sending recites mere data transmission, which is considered insignificant extra-solution activity – see MPEP 2106.05(g)),
logging a record of the resulting action into a memory; (In step 2A, prong 2, logging recites mere data receiving, which is considered insignificant extra-solution activity – see MPEP 2106.05(g)),
and updating, by the processor, the set of metrics based on the resulting action to further train the model, (Mere instructions to apply an exception using generic computer – see MPEP 2106.05(f)),
Since the claim as a whole, looking at the additional elements individually and in combination, does not contain any other additional elements that are indicative of integration into a practical application, the claim is “directed” to an abstract idea.
In step 2B of the 101-analysis set forth in the 2019 PEG, the examiner has determined that the claim does not include additional elements that are sufficient to amount to significantly more than the judicial exception.
As discussed above, additional elements iv and v recite generic computer components being used as a tool. Additional elements vi and ix recite mere instructions to apply the judicial exception using generic computer components, which are not indicative of significantly more. The additional elements vii and viii recite mere data gathering or transmission, and are considered insignificant extra-solution activities. In step 2B, these insignificant extra-solution activities are well understood routine and conventional activities, which include receiving or transmitting data over a network from court case Symantec, 838 F.3d at 1321, 120 USPQ2d at 1362 (utilizing an intermediary computer to forward information); TLI Communications LLC v. AV Auto. LLC, 823 F.3d 607, 610, 118 USPQ2d 1744, 1745 (Fed. Cir. 2016), – see MPEP 2106.05(d) (II)(i)),
Considering the additional elements individually and in combination, and the claim as a whole, the additional elements do not provide significantly more than the abstract idea. Therefore, the claim is not patent eligible.
Claim Rejections - 35 USC § 103
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 (i.e., changing from AIA to pre-AIA ) 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.
The following is a quotation of 35 U.S.C. 103 which forms the basis for all obviousness rejections set forth in this Office action:
A patent for a claimed invention may not be obtained, notwithstanding that the claimed invention is not identically disclosed as set forth in section 102, if the differences between the claimed invention and the prior art are such that the claimed invention as a whole would have been obvious before the effective filing date of the claimed invention to a person having ordinary skill in the art to which the claimed invention pertains. Patentability shall not be negated by the manner in which the invention was made.
The factual inquiries for establishing a background for determining obviousness under 35 U.S.C. 103 are summarized as follows:
1. Determining the scope and contents of the prior art.
2. Ascertaining the differences between the prior art and the claims at issue.
3. Resolving the level of ordinary skill in the pertinent art.
4. Considering objective evidence present in the application indicating obviousness or nonobviousness.
Claims 1, 2, 3, 11, 12, 13 and 21, are rejected under 35 U.S.C. 103 over Bao, X. et al. in US PG Pub. No. US20230068465A1, published on March 2, 2023, (hereafter, Bao), in view of Perry, M. in US PG Pub. No. US20200204514A1, published on June 25, 2020, (hereafter, Perry), further in view of Ike, C. et al., in “Advancing machine learning frameworks for customer retention and propensity modeling in ecommerce platforms,” available on https://www.researchgate.net/profile/Olukunle-Amoo/publication/386893780_Advancing_machine_learning_frameworks_for_customer_retention_and_propensity_modeling_in_E-Commerce_Platforms/links/677994fc894c55208542f53a/Advancing-machine-learning-frameworks-for-customer-retention-and-propensity-modeling-in-E-Commerce-Platforms.pdf , published on 24 February 2023, (hereafter, Ike).
Claim 1:
Regarding claim 1, Bao teaches “A method of interacting with a customer on an e-commerce platform, the method comprising: … evaluating, by the processor, whether the customer would respond to one out of the plurality of messages;”
See Bao in [0075] describe “A marketing susceptibility decision engine (MSDE) 375 may, in some embodiments, implement the same or different ML/AI derived algorithms to analyze the customer action information 60. The MSDE 375 utilizes a different set of the customer action information 60 (different than what the PTBDE 365 or NBADE 370 utilizes) to identify how likely a customer is to change their purchasing behavior based on viewing advertisements/promotions. Thus the output of the MSDE 375 may be in the form of a marketing susceptibility list 378. For example, the MSDE 375 uses information specific to a particular customer (e.g., the customer's history of new purchases following a promotion, history of products that the customer searched for on the external front end system 103 following a promotion, time spent viewing promotions on the external front end system 103) to predict whether a customer will be likely to respond to advertisements/promotions in general or for specific products/product categories by making a purchase. For example, the marketing susceptibility list 378 may indicate an estimated likelihood of change in purchasing behavior for product categories based on receiving marketing for each of the product categories. This prediction by the MSDE 375 can be made, for example, via deep learning machine learning methods.” The examiner construes messages to include any notifications that direct a customer with information about the e-commerce platform, which includes advertisements or promotions for certain products. The customer’s response includes an action such as making a purchase, and shows responding to a message. Here, Bao shows using a marketing susceptibility decision engine unit 375 to predict if a customer responds to advertisements or not.
Further, see Bao in [0008] describe “Yet another aspect of the present disclosure is directed to a system for targeted advertising to a customer including: memory storing instructions; and at least one processor configured to execute the instructions to: receive customer action information associated with the customer; receive a plurality of advertising campaigns;” Here, Bao describes a processor used in the system.
Further, Bao teaches “recommending, by the processor, at least one out of the plurality of messages that results in the highest probability of response from the customer;”
See Bao in [0076] describe “The ranked list of advertising campaigns 395a-n may be in any suitable form (plain text file, spreadsheet, etc.) that indicates that certain advertising campaigns 395 are prioritized for a customer over others. For example, a propensity table generated by the PTBDE 365 may include columns for member identification, product category, and a propensity score. Additionally, an NBA table generated by the NBADE 370 may include columns for member identification, and NBA impact. Business rules 335 may merge the propensity table with the NBA table and then label and rank each member identification on propensity and NBA output by product (example group labels may include core group, nominal group, and no signal). In some arrangements the list 359 may include as few as one advertising campaign 395a. In arrangements where there are multiple advertising campaigns 395a-n are chosen by the CPDE 390, not all of the campaigns 395a-n may be presented to the customer at the customer device 102... In arrangements where there are multiple advertising campaigns 395 a-n are chosen by the CPDE 390, not all of the campaigns 395 a-n may be presented to the customer at the customer device 102. For example, if the CPDF 390 chooses a diaper promotion as the top ranked advertising campaign 395 a, a running shoe promotion as the second ranked advertising campaign 395 b, and a jewelry promotion as the third ranked advertising campaign 395 c, then only the top ranked (i.e. most likely to maximize lifetime revenue from the customer) campaign 395 a may be initially presented to the customer at the device 102 when logged onto the system 103.” Here, Bao mentions a ranking of a list of advertisements, and the system only presents the top ranked advertised campaign to the customer shows the system (which includes a processor), recommending at least one out of the plurality of messages. The highest probability of response is construed by examiner to mean increased chance of a customer to take an action within the e-commerce platform such as to maximize lifetime revenue.
Also, see Bao in [0071-0072] note “As seen in FIG. 4 , the data lake 350 contains a collection of customer action information 60. In some embodiments, the internal front end system 105 updates this information periodically, in real-time, or on demand, for example, by retrieving relevant data from data lake 350. Periodically, certain sets of the customer action information 60 are directed by the processor 320 to be evaluated under several machine learning (ML)/artificial intelligence (AI) derived algorithms implemented by various decision engines 365, 370, 375. The ML/AI methods include deep learning and random forest. It is noted that decision engines 365, 370, 375 are exemplary and that less or more than these may be used by the processor 320. [0072] The propensity to buy decision engine (PTBDE) 365 implements one of these ML/AI derived algorithms. The PTBDE 365 utilizes some of the customer action information 60 to identify customer interest in different sets of products over a set period of time. In some embodiments, the PTBDE 365 is configured to determine which customers have a high propensity to buy a particular target category/brand/device, and therefore drive a higher incremental purchase rate.” Bao in [0071-0072] connects customer action with purchase rate which describes maximizing revenue from [0076] as a form of response by the customer. Further, see Bao in [0008] describe a processor similar to limitation above.
Further, Bao teaches “sending, by the processor, the recommended one out of the plurality of messages to the customer;”
See Bao in [0076] describe “In some arrangements the list 359 may include as few as one advertising campaign 395a. In arrangements where there are multiple advertising campaigns 395a-n are chosen by the CPDE 390, not all of the campaigns 395a-n may be presented to the customer at the customer device 102…While the above-described three different decision engines 365, 370, 375 and their associated ML/AI derived algorithms are described above, this list is not exhaustive and the processor 320 may make use of other types of decision engines or other ML/AI derived algorithms. Outputs 368, 373, 378 of the above-described decision engines 365, 370, 375 along with the plurality of advertising campaigns 80 are fed into the customer promotion decision engine (CPDE) 390. The processor 320 running the CPDE 390 uses various business rules 355 to generate a ranked list of advertising campaigns 395a-n selected from the plurality of advertising campaigns 80 to be offered to the customer at device 102.” Here, Bao mentions the processor unit 320 that runs the list of advertising campaigns which the “list 359 may include as few as one advertising campaign 395a,” and may select one of those advertisements (i.e. recommended one), is offered (i.e. sent) to the customer via device unit 102. See Bao in [0074] for details.
Further, Bao teaches “determining, by the processor,…”, and “updating, by the processor, …”
See Bao in [0008] describe “Yet another aspect of the present disclosure is directed to a system for targeted advertising to a customer including: memory storing instructions; and at least one processor configured to execute the instructions to: receive customer action information associated with the customer; receive a plurality of advertising campaigns;” Here, Bao describes a processor used in the system that runs instructions, including determining or updating steps.
However, Bao did not teach “…providing, by a processor, a model to rank a plurality of messages based on a set of metrics;” or “determining, by the processor, a resulting action from the customer; and updating, by the processor, the set of metrics based on the resulting action to further train the model.”
In an analogous art, Perry teaches “…providing, by a processor, a model to rank a plurality of messages based on a set of metrics;”
See Perry in [0083] describe “In embodiments, the prioritization of messages may be based on attributes or a profile of the sender of the message, such as determined from the content of the message or based on other information. In embodiments, the profile of a user may take into account the past browsing history of a sender and how much time the sender has spent viewing the merchant's products and services.” Here, Perry shows that prioritization or ranking of messages is based on user profiles of the message, which is part of a set of metrics. From the specification [0015], the “ set of metrics contains information about the e-commerce platform, information about the customer; and information about the customer browsing session.” See Perry in [0081-0082,0084-0087] for more details.
Further, see Perry in [0078] mention "In order to prioritize messages coming across all incoming message communication methods, the device responsible for analyzing and ordering messages, as described herein, uses algorithm inputs to rank message and/or conversation priority and then presents the message and/or conversation to the merchant or merchant users in a prioritized fashion," Perry describes ranking messages by a priority using algorithm inputs, and described from [0083] that ranking is based on a set of metrics.
It would have been obvious for one of ordinary skill in the art before the effective filing date of the claimed invention to combine the base reference of Bao and incorporate into the teachings of Perry because both references teach a model to rank messages based on a set of metrics.
One of ordinary skill in the art would be motivated to do so because “a merchant may employ all or any combination of these, such as maintaining a business through a physical storefront utilizing POS devices 152, maintaining a virtual storefront through the online store 138, and utilizing a communication facility 129 to leverage customer interactions and analytics 132 to improve the probability of sales,” (see Perry in [0039]).
However, Bao in view of Perry, did not teach “determining, by the processor, a resulting action from the customer; and updating, by the processor, the set of metrics based on the resulting action to further train the model.”
In an analogous art, Ike teaches “determining, by the processor, a resulting action from the customer;”
See Ike in page 193, section 2. Importance of Customer Retention and Propensity Modeling describe "By leveraging propensity modeling, businesses can predict which customers are most likely to respond positively to specific promotions or products, thereby enhancing the effectiveness of marketing efforts. Instead of casting a wide net with generalized messages, companies can craft campaigns that are finely tuned to customer needs and preferences, thereby increasing engagement and fostering loyalty.” Ike describes here that using a method called propensity modeling, this can help identify whether customers are likely to react and positively reply to specific products (i.e. determine a resulting action from the customer).
See Ike also in page 193 describe in first paragraph “Propensity modeling is a predictive analytics technique used to estimate the likelihood of a customer taking a specific action based on historical data. These actions could include making a purchase, subscribing to a service, or, conversely, churning or abandoning a cart. The primary objective of propensity modeling is to assist businesses in identifying high value customers and anticipating their future behavior, enabling marketers to create targeted strategies that influence customer actions positively (Zulaikha et al., 2020). In e-commerce, propensity models can be applied to various business scenarios. For instance, predicting the likelihood of a customer making a purchase allows businesses to target individuals with personalized offers or promotions.” Here, Ike illustrates the resulting actions that a customer could take.
Further, Ike teaches “and updating, by the processor, the set of metrics based on the resulting action to further train the model.”
See Ike in page 197 in section 2.4 Challenges in Implementation, describe “Another significant challenge is managing the scalability of ML models in the context of large-scale e-commerce operations. E-commerce businesses often deal with enormous datasets, including millions of customer interactions, purchases, and browsing histories, as well as real-time transactional data. The challenge is not just the volume of data, but also its velocity and variety, with new data being generated continuously from various sources such as mobile apps, websites, and social media platforms. To address scalability, e-commerce companies must develop infrastructure capable of handling these large-scale datasets efficiently. Traditional data processing systems may not be sufficient, necessitating the use of big data technologies and distributed computing frameworks such as Hadoop or Apache Spark (Osman, 2019). These systems enable the parallel processing of large datasets and facilitate real-time data analytics, which is critical for updating propensity models with fresh customer behavior data. Additionally, scalable machine learning frameworks, such as TensorFlow and PyTorch, must be leveraged to train and deploy models on massive datasets without compromising performance. Achieving scalability requires careful design and investment in cloud based infrastructure, which can be costly for smaller organizations but is essential for handling the dynamic and expanding nature of e-commerce data.” Update customer behavior data, which is considered a metric, and is used to train and run models based on the updated information.
See Ike in page 197, section 2.4, paragraph 3 “In the context of customer retention, e-commerce businesses need to be able to understand the factors that influence the model’s predictions. For example, a propensity model may predict that a specific customer is likely to churn, but it may not be clear which factors (e.g., recent purchases, frequency of visits, or customer service interactions) contributed to the prediction.”
It would have been obvious to one of ordinary skill in the art before the effective filing date of the claimed invention to combine the reference of Bao along with the reference of Perry with the teachings of Ike by using the teachings of Bao and Perry of ranking messages based on a set of metrics to build a model, with Ike’s teaching of updating the set of metrics based on the resulting action to further train the model, and determine a resulting action from a customer.
One of ordinary skill in the art would be motivated to do so because by integrating Ike’s framework into the methods of Bao and Perry, one with ordinary skill in the art would achieve the goal of providing “successful implementations, including case studies from leading e-commerce platforms, demonstrate the potential of ML to improve customer engagement, increase lifetime value, and drive business growth. Looking forward, integrating AI and ML for enhanced personalization, leveraging real-time predictive analytics, and addressing ethical
considerations like bias and fairness are crucial for advancing these frameworks,” and “Customer retention is vital, as retaining an existing customer is significantly more cost-effective than acquiring a new one. Propensity modeling further complements retention strategies by predicting customer actions, such as purchases …, enabling businesses to tailor their marketing efforts effectively” for customers (see Ike in abstract in page 191).
Claim 2:
Regarding claim 2, Bao in view of Perry, further in view of Ike, teach the limitations of claim 1. Further, Bao teaches "2. The method of claim 1, wherein said model is a machine learning model."
See Bao in [0008] "Yet another aspect of the present disclosure is directed to a system for targeted advertising to a customer including: memory storing instructions; and at least one processor configured to execute the instructions to: receive customer action information associated with the customer; receive a plurality of advertising campaigns; generate, using a first algorithm based on artificial intelligence and/or machine learning models, a list of products derived from the customer action information that the customer may have interest in over a first time period; generate, using a second algorithm, a ranked list of product categories derived from the customer action information that, if purchased by the customer, would generate a highest amount of revenue over a second time period, the second algorithm based on artificial intelligence and/or machine learning models;"
Also, see Bao in [0072] describe " For example, the PTBDE 365 uses information specific to a particular customer (e.g., the customer's previous purchases, products that the customer search for on the external front end system 103, and other browsing history on the system 103) to predict a list of products that customer would be interested in purchasing over the next seven days. The list 368 may or may not be ranked in order of estimated customer interest. The prediction by the PTBDE 365 can be made, for example, via deep learning and/or random forest machine learning methods that use multiple layers to progressively extract higher-level features from the raw input so that predictions made by the PTBDE 365 will be more accurate over time." Here, Bao shows using machine learning models, such as random forest or deep learning to analyze customer actions.
Claim 3:
Regarding claim 3, Bao in view of Perry, further in view of Ike, teach the limitations of claim 2.
Further, Perry teaches “The method of claim 2, wherein the set of metrics contains information about the e-commerce platform, information about the customer; and information about the customer browsing session.”
See Perry in paragraph [0037] describe “While the disclosure throughout contemplates that a ‘merchant’ and a ‘customer’ may be more than individuals, for simplicity the description herein may generally refer to merchants and customers as such. All references to merchants and customers throughout this disclosure should also be understood to be references to groups of individuals, companies, corporations, computing entities, and the like, and may represent for-profit or not-for-profit exchange of products. Further, while the disclosure throughout refers to ‘merchants’ and ‘customers’, and describes their roles as such, the e-commerce platform 100 should be understood to more generally support users in an e-commerce environment, and all references to merchants and customers throughout this disclosure should also be understood to be references to users, such as where a user is a merchant-user (e.g., a seller, retailer, wholesaler, or provider of products), a customer-user (e.g., a buyer, purchase agent, or user of products), a prospective user (e.g., a user browsing and not yet committed to a purchase, a user evaluating the e-commerce platform 100 for potential use in marketing and selling products, and the like), a service provider user (e.g., a shipping provider 112, a financial provider, and the like), a company or corporate user (e.g., a company representative for purchase, sales, or use of products; an enterprise user; a customer relations or customer management agent, and the like), an information technology user, a computing entity user (e.g., a computing bot for purchase, sales, or use of products), and the like.” Here, Perry describes information related to the e-commerce platform, which helps support customer users, and service providers, and related interactions.
Further, see Perry in [0038] describe “The e-commerce platform 100 may provide a centralized system for providing merchants with online resources and facilities for managing their business. The facilities described herein may be deployed in part or in whole through a machine that executes computer software, modules, program codes, and/or instructions on one or more processors which may be part of or external to the platform 100. Merchants may utilize the e-commerce platform 100 for managing commerce with customers, such as by implementing an e-commerce experience with customers through an online store 138, through channels 110A-B, through POS devices 152 in physical locations (e.g., a physical storefront or other location such as through a kiosk, terminal, reader, printer, 3D printer, and the like), by managing their business through the e-commerce platform 100, and by interacting with customers through a communications facility 129 of the e-commerce platform 100, or any combination”, Here, Perry describes information about the e-commerce platform, and in [0037] describe the various types of users that will interact with this platform.
Further, see Perry in [0083] note “In embodiments, the prioritization of messages may be based on attributes or a profile of the sender of the message, such as determined from the content of the message or based on other information. In embodiments, the profile of a user may take into account the past browsing history of a sender and how much time the sender has spent viewing the merchant's products and services.” Here, Perry describes user information includes a browsing history of the sender or customer.
It would have been obvious for one of ordinary skill in the art before the effective filing date of the claimed invention to combine the base reference of Bao and incorporate into the teachings of Perry because both references teach a model to rank messages based on a set of metrics.
One of ordinary skill in the art would be motivated to do so because “a merchant may employ all or any combination of these, such as maintaining a business through a physical storefront utilizing POS devices 152, maintaining a virtual storefront through the online store 138, and utilizing a communication facility 129 to leverage customer interactions and analytics 132 to improve the probability of sales,” (see Perry in [0039]).
Claim 11:
Regarding claim 11, Bao teaches “11. An apparatus, comprising: a memory for storing instructions; and a processor configured to execute the instructions and thereby cause the apparatus to at least:..”
See Bao in [0007] describe “Another aspect of the present disclosure is directed to a system for targeted advertising to a customer including: a memory storing instructions; and at least one processor configured to execute the instructions to: receive customer action information associated with the customer; receive a plurality of advertising campaigns; generate, using a first algorithm, a list of products derived from the customer action information that the customer may have interest in over a first time period;” Bao here shows using a system, (i.e. apparatus) that includes a memory and a processor to run instructions in receiving customer actions and creating a list of products the customer may be interested in purchasing using an algorithm.
Further, Perry teaches “send, via an interface, …; said interface being adapted to display the message;”
See Perry describe in [0071] “In an embodiment, the present disclosure relates generally to the receipt, prioritizing and reply to electronic messages within the e-commerce platform 100. In embodiments, the e-commerce platform 100 includes a communications component 129 that, in embodiments, may communicate directly through the commerce management engine 136. In alternative embodiments, the communication component 129 operates as an internal or external application 142A-B or otherwise communicates through an interface 140A-B. In alternative embodiments, the communication component 129 operates to enable customers and merchants located remotely from each other to communicate through a network, such as an electronic network (for example, a wide area network).” Also, see Perry in [0091] describe “In another embodiment, customer factors and message content are both used to prioritize messages or display messages in an enhanced display.” Here, Perry mentions using an interface to allow customers and merchants send and also display messages.
Further, see Perry in [0048] mention “The e-commerce platform 100 may provide for a communications facility 129 and associated merchant interface for providing electronic communications and marketing, such as utilizing an electronic messaging aggregation facility for collecting and analyzing communication interactions between merchants, customers, merchant devices 102, customer devices 150, POS devices 152, and the like, to aggregate and analyze the communications, such as for increasing the potential for providing a sale of a product, and the like.
It would have been obvious for one of ordinary skill in the art before the effective filing date of the claimed invention to combine the base reference of Bao and incorporate into the teachings of Perry because both references teach a model to rank messages based on a set of metrics.
One of ordinary skill in the art would be motivated to do so because “a merchant may employ all or any combination of these, such as maintaining a business through a physical storefront utilizing POS devices 152, maintaining a virtual storefront through the online store 138, and utilizing a communication facility 129 to leverage customer interactions and analytics 132 to improve the probability of sales,” (see Perry in [0039]).
Regarding additional limitations of claim 11, the claim recites similar limitations as corresponding independent claim 1 and is rejected for similar reasons as claim 1 using similar teachings and rationale.
Claims 12 - 13:
Regarding claims 12 and 13, the claims recite similar limitations as corresponding claims 2 and 3 and are rejected for similar reasons as claims 2 and 3 using similar teachings and rationale.
Claim 21:
Regarding claim 21, Bao teaches “A method of interacting with a customer on an e-commerce platform, the method comprising: … generating, by a processor, a plurality of messages based on a set of metrics associated with the customer;”
See Bao in [0076] mention “The processor 320 running the CPDE 390 uses various business rules 355 to generate a ranked list of advertising campaigns 395 a-n selected from the plurality of advertising campaigns 80 to be offered to the customer at device 102.” Further, see Bao in [0075] describe “Thus the output of the MSDE 375 may be in the form of a marketing susceptibility list 378. For example, the MSDE 375 uses information specific to a particular customer (e.g., the customer's history of new purchases following a promotion, history of products that the customer searched for on the external front end system 103 following a promotion, time spent viewing promotions on the external front end system 103) to predict whether a customer will be likely to respond to advertisements/promotions in general or for specific products/product categories by making a purchase.” Here, Bao mentions generating a list of advertising campaigns, which contain messages based on a customer’s history of purchases following a promotion or a customer’s profile of previously searched products at the e-commerce platform (customer profile relates to information about the customer which is part of the set of metrics, see specification in [0011] discuss “the set of metrics contains information about the e-commerce platform, information about the customer; and information about the customer browsing session.”). Examiner construes messages to mean any notifications or information, from any source that delivers information to customers, and includes advertisements.
Later, see Bao in [0082] mention “Step 410 is sending a first communication associated with a first advertising campaign 195 a of the plurality of advertising campaigns 80 to a customer device 102, the first advertising campaign 195 a chosen by a third algorithm 390 that incorporates, as input, the list of products 168 and the ranked list of product categories 173. For example, based on business rules 355, the CPDE 390 may determine that the best promotion to offer the customer is a promotion for dog food since pet food is the determined highest revenue generating product category and the customer has shown a propensity to purchase dog food within the next week.” Bao mentions that the system CPDE 390 may determine the best promotion (i.e. message) to offer the customer is a promotion for dog food since based on the customer’s profile, this customer has shown a pattern to purchase dog food before (i.e. based on a set of metrics associated with the customer).
Further, Ike teaches “… logging a record of the resulting action into a memory;”
See Ike in page 198, section 2.5 Case studies and applications describe “Sephora, an e-commerce leader in beauty products, has integrated ML for customer retention through personalized experiences both online and in-store. Their use of AI powered tools, such as a virtual artist, allows customers to try out different makeup looks and see personalized product recommendations based on their preferences. By combining this with predictive models that analyze purchase patterns and customer behavior, Sephora has successfully improved customer engagement and retention (Kumari et al., 2020). Their recommendation engine, powered by machine learning, optimizes product suggestions, which leads to increased customer satisfaction and repeat business.” Here, Ike describes an e-commerce platform that uses models to track purchase patterns and customer behavior. Resulting action is construed by examiner to mean any behaviors of customers interacting in a commercial setting.
Further, see Ike in page 197, section 2.4 Challenges in implementation mention “These regulations require companies to ensure that customer data is collected, stored, and processed transparently and securely, with explicit consent from customers.” This shows that Ike shows all customer data is collected and stored in the memory of systems from e-commerce platforms.
It would have been obvious to one of ordinary skill in the art before the effective filing date of the claimed invention to combine the reference of Bao along with the reference of Perry with the teachings of Ike by using the teachings of Bao and Perry of ranking messages based on a set of metrics to build a model, with Ike’s teaching of updating the set of metrics based on the resulting action to further train the model, and determine a resulting action from a customer.
One of ordinary skill in the art would be motivated to do so because by integrating Ike’s framework into the methods of Bao and Perry, one with ordinary skill in the art would achieve the goal of providing “Successful implementations, including case studies from leading e-commerce platforms, demonstrate the potential of ML to improve customer engagement, increase lifetime value, and drive business growth. Looking forward, integrating AI and ML for enhanced personalization, leveraging real-time predictive analytics, and addressing ethical
considerations like bias and fairness are crucial for advancing these frameworks,” and “Customer retention is vital, as retaining an existing customer is significantly more cost-effective than acquiring a new one. Propensity modeling further complements retention strategies by predicting customer actions, such as purchases …, enabling businesses to tailor their marketing efforts effectively” for customers (see Ike in abstract in page 191).
Regarding additional limitations of claim 21, the claim recites similar limitations as corresponding independent claim 1 and is rejected for similar reasons as claim 1 using similar teachings and rationale.
Claims 4 and 14 are rejected under 35 U.S.C. 103 over Bao, in view of Perry, further in view of Ike, and further in view of Dudik, M. et al. from US PG Pub. No. US20200394473A1, published on December 17, 2020, (hereafter, Dudik).
Claim 4:
Regarding claim 4, Bao in view of Perry, further in view of Ike, teach the limitations of claim 3.
However, Bao in view of Perry, further in view of Ike, did not teach “The method of claim 3, wherein the training of the machine learning model is completed using at least one of: inverse propensity-scoring algorithm; doubly robust algorithm; and importance weighted regression algorithm.”
In an analogous art, Dudik teaches “The method of claim 3, wherein the training of the machine learning model is completed using at least one of: inverse propensity-scoring algorithm; doubly robust algorithm; and importance weighted regression algorithm.”
See Dudik in [0021-0022] note “In examples, doubly robust estimation uses a combination of inverse propensity scoring and direct modeling, where inverse propensity scoring may be used to correct a distribution mismatch by reweighting the data, while direct modeling may be used to reduce the impact of large importance weights. As an example, direct modeling comprises generating and using a regression model to predict rewards. In other examples, a reward predictor may be trained and used to generate a predicted reward for a given context… Doubly robust estimation may yield results that are less biased or unbiased, and that have a smaller variance than those achieved using inverse propensity scoring. Further, doubly robust estimation may be asymptotically optimal under weaker assumptions than direct modeling. However, since doubly robust estimation may use the same importance weights as inverse propensity scoring, its variance can still be high, unless the reward predictor is highly accurate. Accordingly, in some examples, doubly robust estimation may be further improved according to aspects described herein by clipping or removing large importance weights, such as by using either a quality-agnostic estimator or a quality-based estimator to weight reward predictions used by a doubly robust estimation model. While weight clipping or shrinking may incur a small bias, it may also substantially decrease the variance, which can result in a lower mean squared error (MSE) than implementing doubly robust estimation without such techniques. This disclosure presents systems and methods that improve off-policy evaluation through weight clipping.”
Further, see Dudik in [0038] “At operation 204, a reward predictor is generated. As described above, a reward predictor may generate an expected reward. For example, the reward predictor generates an expected reward using the historical data associated with the logging policy, as was accessed at operation 202. Thus, in examples, the reward predictor generates the predicted reward given the context and the associated action that was determined by the logging policy based on the context. In some examples, the reward predictor may be modelled as a regression function (e.g., as may be the case when using a direct modeling approach).” Dudik describes using doubly robust estimation, which also uses importance weights similar to inverse propensity scoring from “since doubly robust estimation may use the same importance weights as inverse propensity scoring”, and can be modelled in regression function. In [0021], Dudik also mentions incorporates all three methods into training of algorithms or models. Also see Dudik in [0039] for more details.
It would have been obvious to one of ordinary skill in the art before the effective filing date of the claimed invention to combine the references of Bao, Perry, and Ike by using the teachings of Bao, Perry, and Ike of ranking messages based on a set of metrics to build a model, and updating metrics to train that model, and incorporate with Dudik’s teaching of using methods of inverse propensity-scoring algorithm; doubly robust algorithm; and importance weighted regression algorithm to train the model.
One of ordinary skill in the art would be motivated to do so because by integrating Dudik’s framework into the methods of Bao, Perry, and Ike, one with ordinary skill in the art would achieve the goal of providing “aspects described herein allow for better evaluation and optimization of policies. For example, less historical data associated with a logging policy may be required to evaluate a target policy, thereby reducing computational overhead associated with acquiring and processing the historical data. Additionally, as a result of efficiently using less historical data for off-policy evaluation, the complexities of obtaining relevant and current historical data are reduced, thereby minimizing the impact of potentially stale data on the evaluation,” (see Dudik in [0023]), and “accordingly, the techniques described herein achieve a similar or improved result as compared to previous solutions. Additionally, such off-policy evaluation techniques may exhibit improved finite-sample performance and may therefore achieve results that are comparable to other techniques using comparatively less historical data,” (see Dudik in [0028]).
Claim 14:
Regarding claim 14, the claim recites similar limitations as corresponding claim 4, and is rejected for similar reasons as claim 4 using similar teachings and rationale.
Claims 5 and 15 are rejected under 35 U.S.C. 103 over Bao, in view of Perry, further in view of Ike, further in view of Dudik, and further in view of Navarro, L. in “Strategic integration of content analytics in content marketing to enhance data-informed decision making and campaign effectiveness”, published on July 4, 2017, available at https://www.researchgate.net/profile/Laura-Fernanda-Malagon-Navarro/publication/387130254_Strategic_Integration_of_Content_Analytics_in_Content_Marketing_to_Enhance_Data-Informed_Decision_Making_and_Campaign_Effectiveness/links/676179a62d60b863e276ba10/Strategic-Integration-of-Content-Analytics-in-Content-Marketing-to-Enhance-Data-Informed-Decision-Making-and-Campaign-Effectiveness.pdf in the Journal of Artificial Intelligence and Machine Learning in Management, (hereafter, Navarro).
Claim 5:
Regarding claim 5, Bao in view of Perry, further in view of Ike, and further in view of Dudik, teach the limitations of claim 4.
However, Bao in view of Perry, further in view of Ike, further in view of Dudik, did not teach “The method of claim 4, wherein the method further comprises the step of amending at least one of the plurality of messages if the message receives a low probability of response from the customer.”
In an analogous field, Navarro teaches “The method of claim 4, wherein the method further comprises the step of amending at least one of the plurality of messages if the message receives a low probability of response from the customer.”
See Navarro in page 8, section “Machine Learning for Predictive Modeling”, first and fourth paragraphs of section, mention “These predictive models allow for anticipatory adjustments in content strategy by analyzing how specific content features—such as structure, tone, media type, and thematic elements—correlate with user interaction patterns. Through such insights, marketers can forecast user responses more accurately and tailor content to maximize engagement… Beyond linear and ensemble methods, advanced machine learning techniques like gradient boosting and neural networks capture nonlinear relationships between content features and audience responses. Gradient boosting iteratively improves prediction accuracy by prioritizing misclassified or mispredicted instances in previous iterations, thereby refining the model’s sensitivity to subtle patterns within the data. This approach is particularly beneficial for content analytics, where variations in user engagement may be influenced by minor adjustments in tone or formatting. Gradient boosting can therefore detect and leverage these nuances, enhancing predictive precision for complex, multidimensional data.” Navarro mentions here of adjusting (i.e. amending) content features such as elements of structure, tone, or type of media in delivering the message, showing the action part of adjusting or amending the message content to resolve low customer response rates to messages.
Later, see Navarro in page 14, fourth paragraph describe “For example, if a particular social media channel exhibits lower engagement rates than expected, resources can be reallocated to more effective channels or content formats.” Navarro mentions the lower engagement rates relate to low probability of response from the customer. This reallocation is the action taken from page 8, where Navarro mentions if a content of a particular message receives low customer engagement, then that content of the message, such as tone, structure or other format, gets edited by a predictive method to change user engagement.
It would have been obvious to one of ordinary skill in the art before the effective filing date of the claimed invention to combine the references of Bao, Perry, Ike, and Dudik by using the teachings of Bao, Perry, Ike, and Dudik, of ranking messages based on a set of metrics to build a model, and updating metrics to train that model, and incorporate with Navarro’s teaching of amending at least one of the messages if that message receives a low response rate from a customer.
One of ordinary skill in the art would be motivated to do so because by integrating Navarro’s framework into the methods of Bao, Perry, Ike, and Dudik, one with ordinary skill in the art would achieve the goal of providing methods for “audience targeting and personalization, guided by segmentation insights, therefore enable a precision-based approach to content distribution, fostering greater resonance between the content and its intended audience. This approach ultimately strengthens the effectiveness of content campaigns by aligning strategic objectives with the preferences and behaviors of distinct user groups, resulting in optimized engagement and increased conversion potential,” (see Navarro in page 11, col. 2, second paragraph, part of section Audience targeting and personalization).
Claim 15:
Regarding claim 15, the claim recites similar limitations as corresponding claim 5, and is rejected for similar reasons as claim 5 using similar teachings and rationale.
Claims 6, 7, 8, 16, 17 and 18 are rejected under 35 U.S.C. 103 over Bao, in view of Perry, further in view of Ike, further in view of Dudik, further in view of Navarro, and further in view of Konig, Y. et al., in US PG Pub. No. US20210201327A1, published on July 1, 2021, mentioned in the IDS, (hereafter, Konig).
Claim 6:
Regarding claim 6, Bao in view of Perry, further in view of Ike, further in view of Dudik, further in view of Navarro, teach the limitations of claim 5.
However, Bao in view of Perry, further in view of Ike, further in view of Dudik, further in view of Navarro, did not teach “The method of claim 5, wherein the plurality of messages comprises a text-based prompt; voice-based prompt; image based prompt; or video based prompt.”
In an analogous art, Konig teaches “The method of claim 5, wherein the plurality of messages comprises a text-based prompt; voice-based prompt; image based prompt; or video based prompt.”
See Konig mention in [0038] “By way of background, customer service providers generally offer many types of services through contact centers. Such contact centers may be staffed with employees and/or customer service agents (or simply “agents”), with the agents serving as an interface between an organization, such as a company, enterprise, or government agency, and persons, such as users or customers (hereinafter generally referred to as “customers”). For example, the agents at a contact center may assist customers in making purchasing decisions and receive purchase orders. Similarly, agents may assist or support customers in solving problems with products or services already provided by the organization. Within a contact center, such interactions between contact center agents and outside entities or customers may be conducted over a variety of communication channels, such as, for example, via voice (e.g., telephone calls or voice over IP or VoIP calls), video (e.g., video conferencing), text (e.g., emails and text chat), or through other media.” Here, Konig illustrates various types of communication channels that virtual agents communicate with customers, including text, video, emails, voice formats.
Further, see Konig in paragraph [0125] mention “As an example, a customer may search for keywords spoken by an agent in order to retrieve saved audio that was spoken near the keywords. As another example, after conducting a technical support call, a customer may have the ability to recall and view an image that was shared and annotated by an agent when explaining how to set up a particular piece of equipment. A customer may then be able to view specific details of an interaction, such as the timing of particular spoken lines in the conversation and what files were being shared at the time.” Here, Konig also provides example of an image based conversation that the agent shared with a customer.
Further, see Konig in [0062] mention “According to an embodiment, the media services 249 may provide audio and/or video services to support contact center features such as prompts for an IVR or IMR system (e.g., playback of audio files), hold music, voicemails/single party recordings, multi-party recordings (e.g., of audio and/or video calls), speech recognition, dual tone multi frequency (DTMF) recognition, faxes, audio and video transcoding, secure real-time transport protocol (SRTP), audio conferencing, video conferencing, coaching (e.g., support for a coach to listen in on an interaction between a customer and an agent and for the coach to provide comments to the agent without the customer hearing the comments), call analysis…” Konig here illustrates examples of prompts in various formats, such as audio, video, or calls.
It would have been obvious to one of ordinary skill in the art before the effective filing date of the claimed invention to combine the references of Bao, Perry, Ike, Dudik, and Navarro, by using the teachings of Bao, Perry, Ike, Dudik, and Navarro of using messages based on a set of metrics to build a model, and updating metrics to train that model, and incorporate with Konig’s teaching of messages include either text-based prompt; voice-based prompt; image based prompt; or video based prompt.
One of ordinary skill in the art would be motivated to do so because by integrating Konig’s framework into the methods of Bao, Perry, Ike, Dudik, and Navarro, one with ordinary skill in the art would achieve the goal of providing “the models 252 may include behavior models of customers or agents. The behavior models may be used to predict behaviors of, for example, customers or agents, in a variety of situations, thereby allowing embodiments of the present invention to tailor interactions based on the predictions or to allocate resources in preparation for predicted characteristics of future interactions, and thereby improving overall performance, including improving the customer experience,” (see Konig in [0063] ).
Claim 7:
Regarding claim 7, Bao in view of Perry, further in view of Ike, further in view of Dudik, further in view of Navarro, further in view of Konig, teach the limitations of claim 6.
Further, Perry teaches “7. The method of claim 6, wherein the set of metrics further comprises information about browsing behavior; recurrency; provenance; geolocation; and temporal details.”
See Perry in [0083] describe “In embodiments, the prioritization of messages may be based on attributes or a profile of the sender of the message, such as determined from the content of the message or based on other information. In embodiments, the profile of a user may take into account the past browsing history of a sender and how much time the sender has spent viewing the merchant's products and services.” Here, Perry describes browsing behavior of a sender or customer.
See Perry also describe in paragraphs [0080-0081] “[0080] In embodiments, the prioritization of messages may be based on one or more of many factors (which may also be referred to as data features). Overall, the prioritization logic may be optimizing for the highest probability of making a sale or the highest expected value of a sale (i.e. the probability of making a sale multiplied by the value of the sale). In embodiments, one factor may be customer identity, including whether the customer is a repeat or potential new customer, in general across merchants or with the given merchant. A repeat customer for a given merchant may be more likely to make an additional purchase and so may be prioritized ahead of customers who have not made a previous purchase. Another factor may be the location of the customer; for example, determined through an imputed location through a reverse IP lookup, GPS data, payment data and the like. A nearby customer or a customer in a known wealthy area may be prioritized ahead of others. Another factor may be customer demographics (including customer demographics in relation to typical customers of the products and/or services sold). [0081] Another factor may be a customer's actual or type of previous purchase history; for example, including whether the customer has previously made large or small purchases and the categories of items purchased (such as consumer electronics or cosmetics)...” Perry shows if a customer is a previous customer that was returning to buy items from the platform, and shows the recurrency metric. Perry also shows using GPS data of customer location as a set of metrics (i.e. geographical location), as well as recurrency to show if customer is a repeat customer.
It would have been obvious for one of ordinary skill in the art before the effective filing date of the claimed invention to combine the base reference of Bao and incorporate into the teachings of Perry because both references teach a model to rank messages based on a set of metrics.
One of ordinary skill in the art would be motivated to do so because “a merchant may employ all or any combination of these, such as maintaining a business through a physical storefront utilizing POS devices 152, maintaining a virtual storefront through the online store 138, and utilizing a communication facility 129 to leverage customer interactions and analytics 132 to improve the probability of sales,” (see Perry in [0039]).
Further, Konig teaches “7. The method of claim 6, wherein the set of metrics further comprises information about browsing behavior; recurrency; provenance; geolocation; and temporal details.”
See Konig in [0063] describe” According to an embodiment, the analytics module 250 may provide systems and methods for performing analytics on interaction data from a plurality of different data sources such as different applications associated with a contact center or an organization. Aspects of embodiments of the present invention are also directed to generating, updating, training, and modifying predictors or models 252 based on collected interaction data. The models 252 may include behavior models of customers or agents.” Here, Konig mentions using different data sources, which is considered data provenance. Data provenance is construed to mean tracking a data record’s origin and following that record across time.
Further, see Konig in [0194] note “For the sake of providing examples as to how such interaction predictors may be derived for a given customer, reference will now be made to an exemplary customer “Adam”. To begin the process, the machine learning algorithm of the predictor module 625 may be configured to monitor a given dataset. This dataset may be obtained from any of the several sources of data described herein. For example, one or more data sources may be derived from data maintained within Adam's own customer profile (i.e., customer profile 330). The machine learning algorithm may have access to and monitor several of the types of data stored within Adam's customer profile, e.g., the personal data, interaction data, feedback data, and/or choice data.” Here, Konig mentions using the collected data and tracking updates to a data record such as a customer profile in [0194]. This information includes personal data, interaction data, feedback data, etc. about that customer profile, and shows data provenance.
Further, see Konig in [0125] note “The customer automation system 300 may further document the interaction by storing and indexing any messages, documents, files, and other media involved or shared during in the interaction and, thereby, provide a customer with the ability to later search this material. As an example, a customer may search for keywords spoken by an agent in order to retrieve saved audio that was spoken near the keywords. As another example, after conducting a technical support call, a customer may have the ability to recall and view an image that was shared and annotated by an agent when explaining how to set up a particular piece of equipment. A customer may then be able to view specific details of an interaction, such as the timing of particular spoken lines in the conversation and what files were being shared at the time.” Here, Konig describes temporal details, such as information shared at the time of the conversation, regarding customer interactions. See Konig for further info at [0159] “Returning to the specific example of FIG. 11, the analysis results in identifying target timeframes 530 for each of the pending actions 525,” and at [0154]. See Konig in [0002, 0052-0064] for more information.
It would have been obvious to one of ordinary skill in the art before the effective filing date of the claimed invention to combine the references of Bao, Perry, Ike, Dudik, and Navarro, with Konig, by using the teachings of Bao, Perry, Ike, Dudik, and Navarro of using messages based on a set of metrics to build a model, and updating metrics to train that model, and incorporate with Konig’s teaching of a set of metrics further comprises information about browsing behavior; recurrency; provenance; geolocation; and temporal details.
One of ordinary skill in the art would be motivated to do so because by integrating Konig’s framework into the methods of Bao, Perry, Ike, Dudik, and Navarro, one with ordinary skill in the art would achieve the goal of providing “the models 252 may include behavior models of customers or agents. The behavior models may be used to predict behaviors of, for example, customers or agents, in a variety of situations, thereby allowing embodiments of the present invention to tailor interactions based on the predictions or to allocate resources in preparation for predicted characteristics of future interactions, and thereby improving overall performance, including improving the customer experience,” (see Konig in [0063] ).
Claim 8:
Regarding claim 8, Bao in view of Perry, further in view of Ike, further in view of Dudik, further in view of Navarro, and further in view of Konig, teach the limitations of claim 7.
Further, Konig teaches “The method of claim 7, wherein the resulting action from the customer is a positive action such that the customer accepts the message.”
See Konig in [0082] describe “During the chat conversation, the dialog manager 272 selects a response deemed to be appropriate at the particular point of the conversation flow/script, and outputs the response to the output generator 274. According to an embodiment, the dialog manager 272 may also be configured to compute a confidence level for the selected response and provide the confidence level to the agent device 230. According to an embodiment, every segment, step, or input in a chat communication may have a corresponding list of possible responses. … According to an embodiment, confidence level may be determined based on customer feedback. For example, in response to detecting a negative reaction from a customer to an action or response taken by the chatbot, the confidence level may be reduced. Conversely, in response to detecting a positive reaction from a customer, the confidence level may be increased.” Here, Konig shows that the confidence level relates to a response rate or probability of customer responding, and the positive reaction relates to a positive action.
Further, see Konig in [0174] mention “On the contact center side of the interaction, the present customer profiles also may be used toward improving contact center operations, such as, for example: making call forecasting more context oriented and reliable; improving handle time predictions and queue optimization; improving outbound campaigns (e.g., by targeting customers who are more likely to see value in and respond positively to a particular offer);” Konig mentions that to respond positively to an offer is viewed as a positive action of a customer accepting a message. In this case, the message is conveyed by advertising campaigns.
Further, see Konig in [0136] note "A third identified pending action 525 is referenced as “restore high-speed connection”. In this case, the agent makes statements regarding one or more future actions that will be taken in regard to restoring the customer's high-speed connection once the customer agrees to the premium plan upgrade. When the customer accepts the upgrade, the agent's statements effectively become a promise, and the future actions required to restore the customer's high-speed connection is classified as a pending action." Here, Konig mentions that the customer responds positively relate to a positive action, which in [0136] Konig shows the customer accepts a message regarding an upgrade for a purchased plan.
It would have been obvious to one of ordinary skill in the art before the effective filing date of the claimed invention to combine the references of Bao, Perry, Ike, Dudik, and Navarro, with Konig, by using the teachings of Bao, Perry, Ike, Dudik, and Navarro of using messages based on a set of metrics to build a model, and updating metrics to train that model, and incorporate with Konig’s teaching of a resulting positive action from a customer accepting the message.
One of ordinary skill in the art would be motivated to do so because by integrating Konig’s framework into the methods of Bao, Perry, Ike, Dudik, and Navarro, one with ordinary skill in the art would achieve the goal of providing “the models 252 may include behavior models of customers or agents. The behavior models may be used to predict behaviors of, for example, customers or agents, in a variety of situations, thereby allowing embodiments of the present invention to tailor interactions based on the predictions or to allocate resources in preparation for predicted characteristics of future interactions, and thereby improving overall performance, including improving the customer experience,” (see Konig in [0063] ).
Claim 16:
Regarding claim 16, the claim recites similar limitations as corresponding claim 6, and is rejected for similar reasons as claim 6 using similar teachings and rationale.
Claim 17:
Regarding claim 17, the claim recites similar limitations as corresponding claim 7, and is rejected for similar reasons as claim 7 using similar teachings and rationale.
Claim 18:
Regarding claim 18, the claim recites similar limitations as corresponding claim 8, and is rejected for similar reasons as claim 8 using similar teachings and rationale.
Claims 9, 10, 19, and 20 are rejected under 35 U.S.C. 103 over Bao, in view of Perry, further in view of Ike, further in view of Dudik, further in view of Navarro, further in view of Konig, and further in view of Greystoke, A. et al. in US PG Pub. No. US20150012467A1, published on January 8, 2015, (hereafter, Greystoke).
Claim 9:
Regarding claim 9, Bao in view of Perry, further in view of Ike, further in view of Dudik, further in view of Navarro, and further in view of Konig, teach the limitations of claim 8.
However, Bao in view of Perry, further in view of Ike, further in view of Dudik, further in view of Navarro, and further in view of Konig, did not teach “The method of claim 8, wherein the resulting action from the customer is a negative action such that the customer declines the message.”
In an analogous art, Greystoke teaches “The method of claim 8, wherein the resulting action from the customer is a negative action such that the customer declines the message.”
See also Greystoke describe in paragraph [0091] “ A solution may be chosen from the set of potential results, and the selected or chosen solution can be introduced to the entity group. That entity group may be human or virtual, and the user represented by the entity group may interact with the system to decline the chosen result. The system can receive a returning alert that the result was declined and that another result was selected. Choosing another result can be performed by showing other results, a subset of the results, or the entire results set and by receiving an input indicating a selection of a particular result. Choosing another result may, or may not, result in showing some aspects of the scoring process in order to assist the entities to make another choice. Once another result is selected, it may be referred as a learning set to the engine, which may be a “state of the art” machine learning mechanism such as neural network, genetic algorithm or other current technology that is used in order to tweak the preferences… The small signals may be detected based on the user's interactions and non-interactions, such as adding an item to her/his shopping cart and then either removing the item or allowing the item to sit in the shopping cart for a period of time, such as until a session expires.” Here, Greystoke mentions the user or customer rejecting certain results is recorded as another form of response communicated by the user.
Further, see Greystoke mention in [0363] “In addition, a historical database of items including but not limited to detailed user features and persona, activity, prior expressions of interest, options, previous purchases and previous offer declines is kept. At least the location and time of each of these historical database items is logged and analyzed.” Here, Greystoke mentions that all types of user or customer interactions or communication, such as declines of purchases or rejections of results are recorded in a database. See Greystoke mention more details in [0275].
It would have been obvious to one of ordinary skill in the art before the effective filing date of the claimed invention to combine the references of Bao, Perry, Ike, Dudik, Navarro, and Konig, with Greystoke, by using the teachings of Bao, Perry, Ike, Dudik, Navarro, and Konig, of using messages based on a set of metrics to build a model, and updating metrics to train that model, and incorporate with Greystoke’s teaching of a resulting negative action from a customer declining the message.
One of ordinary skill in the art would be motivated to do so because by integrating Greystoke’s framework into the methods of Bao, Perry, Ike, Dudik, Navarro, and Konig, one with ordinary skill in the art would achieve the goal of providing “improvements to the functioning of a computer by providing enhanced results and dynamic intelligent decisions, thereby creating a specific purpose computer by adding such technology. Thus, the improvements herein provide for technical advantages, such as providing a system in which a user's interaction with a computer system and complex results or decisions are made easier. For example, the systems and processes described herein can be particularly useful to any systems in which a user may want to buy, lease, rent, search, exchange, bid, or barter for goods or services. Further, the improvements herein provide additional technical advantages, such as providing a system in which the personas can operate continuously, apply experiential learning to perform tasks, solve problems, make recommendations, and assist the user by helping manage the user's life experiences to make the user's life easier in terms of dealing with problems, anticipating and solving problems (sometimes before the user is even aware that a problem may exist), managing tasks, and ensuring that all aspects of the user's life receive due attention”, (see Greystoke in [0442] ).
Claim 10:
Regarding claim 10, Bao in view of Perry, further in view of Ike, further in view of Dudik, further in view of Navarro, further in view of Konig, further in view of Greystoke, teach the limitations of claim 9.
Further, Greystoke teaches “ The method of claim 9, wherein the resulting action from the customer is a neutral action such that the customer ignores the message,”
See Greystoke mention in [0275] “If the user would like to run the search again closer to the desired travel date to see if any better results are available, he or she can utilize the price optimization indicator 1655. Once activated, the user can decide when to run the search again, or how frequently to run the search and have the search results or an indicator of new search results sent to him or her. In certain embodiments, the indicator may be sent to the user as an email message, a sms text, an audio alert, a voice message, an image, another type of notification,” Greystoke describes that the results include messages in email, text, audio, or voice format.
Further, see Greystoke describe in paragraph [0091] “ A solution may be chosen from the set of potential results, and the selected or chosen solution can be introduced to the entity group. .. Once another result is selected, it may be referred as a learning set to the engine, which may be a “state of the art” machine learning mechanism such as neural network, genetic algorithm or other current technology that is used in order to tweak the preferences. .. In certain embodiments, the system can also process situations in which a result is presented to a user and the result isn't declined but is ignored by the user or is almost selected by the user, but then the user changes her/his mind. The small signals may be detected based on the user's interactions and non-interactions, such as adding an item to her/his shopping cart and then either removing the item or allowing the item to sit in the shopping cart for a period of time, such as until a session expires.”
Further, see Greystoke in [0120] describe “In certain embodiments, a persona may operate continuously to search for “better” options for a user's query (i.e., better options than that which the user selected, or better options than those that were previously presented to the user). Those better options may be understood to be “better” based on the user's prior interactions with available options (which results he/she viewed, which results he/she viewed for longer than others, and which results he/she ignored).” Here, Greystoke mentions the user or customer ignoring certain results is recorded as another form of response by the user. The examiner construes neutral action to mean a user or customer ignores or not respond to the message or notification sent by a virtual agent system, described from specification in [0013].
It would have been obvious to one of ordinary skill in the art before the effective filing date of the claimed invention to combine the references of Bao, Perry, Ike, Dudik, Navarro, and Konig, with Greystoke, by using the teachings of Bao, Perry, Ike, Dudik, Navarro, and Konig, of using messages based on a set of metrics to build a model, and updating metrics to train that model, and incorporate with Greystoke’s teaching of a resulting neutral action from a customer ignoring the message.
One of ordinary skill in the art would be motivated to do so because by integrating Greystoke’s framework into the methods of Bao, Perry, Ike, Dudik, Navarro, and Konig, one with ordinary skill in the art would achieve the goal of providing “improvements to the functioning of a computer by providing enhanced results and dynamic intelligent decisions, thereby creating a specific purpose computer by adding such technology. Thus, the improvements herein provide for technical advantages, such as providing a system in which a user's interaction with a computer system and complex results or decisions are made easier. For example, the systems and processes described herein can be particularly useful to any systems in which a user may want to buy, lease, rent, search, exchange, bid, or barter for goods or services. Further, the improvements herein provide additional technical advantages, such as providing a system in which the personas can operate continuously, apply experiential learning to perform tasks, solve problems, make recommendations, and assist the user by helping manage the user's life experiences to make the user's life easier in terms of dealing with problems, anticipating and solving problems (sometimes before the user is even aware that a problem may exist), managing tasks, and ensuring that all aspects of the user's life receive due attention”, (see Greystoke in [0442] ).
Claim 19:
Regarding claim 19, the claim recites similar limitations as corresponding claim 9, and is rejected for similar reasons as claim 9 using similar teachings and rationale.
Claim 20:
Regarding claim 20, the claim recites similar limitations as corresponding claim 10, and is rejected for similar reasons as claim 10 using similar teachings and rationale.
Conclusion
Any inquiry concerning this communication or earlier communications from the examiner should be directed to WENWEI ZENG whose telephone number is (571)272-7111. The examiner can normally be reached Monday-Friday, 8am-5pm.
Examiner interviews are available via telephone, in-person, and video conferencing using a USPTO supplied web-based collaboration tool. To schedule an interview, applicant is encouraged to use the USPTO Automated Interview Request (AIR) at http://www.uspto.gov/interviewpractice.
If attempts to reach the examiner by telephone are unsuccessful, the examiner’s supervisor, Usmaan Saeed can be reached at (571) 272-4046. The fax phone number for the organization where this application or proceeding is assigned is 571-273-8300.
Information regarding the status of published or unpublished applications may be obtained from Patent Center. Unpublished application information in Patent Center is available to registered users. To file and manage patent submissions in Patent Center, visit: https://patentcenter.uspto.gov. Visit https://www.uspto.gov/patents/apply/patent-center for more information about Patent Center and https://www.uspto.gov/patents/docx for information about filing in DOCX format. For additional questions, contact the Electronic Business Center (EBC) at 866-217-9197 (toll-free). If you would like assistance from a USPTO Customer Service Representative, call 800-786-9199 (IN USA OR CANADA) or 571-272-1000.
/WenWei Zeng/Examiner, Art Unit 2146
/USMAAN SAEED/Supervisory Patent Examiner, Art Unit 2146
Prosecution Insights