DETAILED ACTION
Summary and Status of Claims
The present application, filed on or after March 16, 2013, is being examined under the first inventor to file provisions of the AIA .
This Office Action is in response to Applicant’s reply filed 8/20/2025.
Claims 21-23 are new.
Claims 1, 5-14, 17-23 are pending.
Claims 1, 5-8, 10-12, 14, 17, and 19 are rejected under 35 U.S.C. 103 as being unpatentable over Datta et al. (US Patent Pub 2018/0113919), in view of Jain et al. (US Patent Pub 2014/0280339) and Dave et al. (US Patent 8,156,073), further in view of Aggarwal et al. (US Patent Pub 2002/0138481).
Claims 9 and 23 are rejected under 35 U.S.C. 103 as being unpatentable over Datta et al. (US Patent Pub 2018/0113919), in view of Jain et al. (US Patent Pub 2014/0280339), Dave et al. (US Patent 8,156,073), and Aggarwal et al. (US Patent Pub 2002/0138481), further in view of Zheng (US Patent Pub 2005/0267809).
Claims 13, 21, and 22 are rejected under 35 U.S.C. 103 as being unpatentable over Datta et al. (US Patent Pub 2018/0113919), in view of Jain et al. (US Patent Pub 2014/0280339), Dave et al. (US Patent 8,156,073), and Aggarwal et al. (US Patent Pub 2002/0138481), further in view of Tavernier (US Patent 10,706,450).
Claim 18 is rejected under 35 U.S.C. 103 as being unpatentable over Datta et al. (US Patent Pub 2018/0113919), in view of Jain et al. (US Patent Pub 2014/0280339), Dave et al. (US Patent 8,156,073), and Aggarwal et al. (US Patent Pub 2002/0138481), further in view of Neal et al. (US Patent Pub 2003/0145277).
Claim 20 is rejected under 35 U.S.C. 103 as being unpatentable over Datta et al. (US Patent Pub 2018/0113919), in view of Jain et al. (US Patent Pub 2014/0280339), Dave et al. (US Patent 8,156,073), and Aggarwal et al. (US Patent Pub 2002/0138481), further in view of Applicant Admitted Prior Art.
The text of those sections of Title 35, U.S. Code not included in this action can be found in a prior Office action.
Note on Prior Art Rejections
In the event the determination of the status of the application as subject to AIA 35 U.S.C. 102 and 103 (or as subject to pre-AIA 35 U.S.C. 102 and 103) is incorrect, any correction of the statutory basis for the rejection will not be considered a new ground of rejection if the prior art relied upon, and the rationale supporting the rejection, would be the same under either status.
Claim Rejections - 35 USC § 103
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 of this title, 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 text of those sections of Title 35, U.S. Code not included in this action can be found in a prior Office action.
The factual inquiries set forth in Graham v. John Deere Co., 383 U.S. 1, 148 USPQ 459 (1966), that are applied 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, 5-8, 10-12, 14, 17, and 19 are rejected under 35 U.S.C. 103 as being unpatentable over Datta et al. (US Patent Pub 2018/0113919) (Datta), of record, in view of Jain et al. (US Patent Pub 2014/0280339) (Jain) of record and Dave et al. (US Patent 8,156,073) (Dave), further in view of Aggarwal et al. (US Patent 2002/0138481) (Aggarwal).
In regards to claim 1, Datta discloses a search system comprising at least one processor (Datta at para. 0087) configured to:
a. search a database, in which a plurality of data items are stored, using a first query entered by a user (Datta at para. 0073)1;
b. provide the user with a data item selected by the user from a search result (Datta at paras. 0042, 0084-85)2;
c. train a learning machine to learn a relationship between a first data item index of the data as an input and the first query as an output (Datta at paras. 0057-60, 0063-64, 0070)3;
d. wherein the first data item index is an attribute of the data item (Datta at para. 0058)4;
e. enter a first target data item index of a target data item that is a new data item to the system into the learning machine and obtain a second query that is output from the learning machine (Datta at paras. 0057-59, 0063-65)5; and
f. register the second query in the database as a second target data item index of the target data item (Datta at para. 0070)6;
g. wherein the second target data item index is not a category of the target data item (Datta at para. 0049)7;
h. wherein the first query and the second query are inaccurate (Datta at para. 0016)8
i. record a combination of the first data item index of the data item selected by the user and a first character information used as the query (Datta at para. 0058)9, and
j. train the learning machine to learn the relationship based on the combination once a predetermined time and date have arrived (Datta at paras. 0049, 0085)10;
l. when the user inputs a new query, provide the user with a first search result based on the first data item index, the second target data item index, and the new query, (Datta at para. 0017)
m. when the user inputs the new query, provide the user with a second search result based on the second target data item index, the first data item index, and the new query (Datta at para. 0017)11;
n. wherein the second target data item index is unique when compared to the first data item index (Datta at paras. 0059, 0063)12;
o. wherein the second target data item index does not contain items of the first data item index (Datta at paras. 0059, 0063)13;
Datta does not expressly disclose wherein the first search result and the second search result are provided on a same screen at a same time. As noted above, Datta does disclose providing a first search result and a second search result in response to a single query. However, Datta does not expressly disclose these results are both provided on a same screen at a same time.
Jain discloses a system and method for providing offers/coupons (i.e., second search result) within search results (i.e., first search result) in response to a query. Jain at paras. 0018-19. The offers/coupons can be determined based on keywords found in the query and user search behavior and intent. Jain at paras. 0042-43. These offers/coupons (i.e., second search result) are displayed along with the regular search results (i.e., first search result) on the same screen at the same time, in response to a single search query. Jain at Fig. 3.
Datta and Jain are analogous art because they are both directed to the same field of endeavor of searching for products of services and including additional related information.
At the time before the effective filing date of the instant Application, it would have been obvious to one of ordinary skill in the art to modify Datta by adding the feature of wherein the first search result and the second search result are provided on a same screen at a same time, as disclosed by Jain.
The motivation to do so would have been because it provides an efficient manner of present to the user relevant coupons/offers without the user providing additional input or performing additional steps. Jain at para. 0019. As discussed above, Datta already discloses providing two search results. Datta also discloses providing results that include advertisements and offers, but does not expressly disclose the results and offers are displayed on the same screen at the same time. Thus, one of ordinary skill in the art, relying on Jain, would have been motivated to modify Datta to display offers and coupons with the search results on the same screen at the same time for the benefit of efficiently providing the user offers without additional steps.
Datta in view of Jain does not expressly disclose wherein the character information entered by the user in the past for a product information is registered as the first data item index.
Dave discloses a method of generating attributes for an item (i.e., registering as the first data item index) utilizing customer queries (i.e., character information entered by the user in the past for a product information). Dave at col. 8, lines 22-36.
Datta, Jain, and Dave are analogous art because they are directed to the same field of endeavor of searching for products and services.
At the time before the effective filing date of the instant Application, it would have been obvious to one of ordinary skill in the art to modify Datta in view of Jain by adding the feature of wherein the character information entered by the user in the past for a product information is registered as the first data item index, as disclosed by Dave.
The motivation for doing so would have been to reduce and eliminate the need for merchants to complete an item summary that identifies attributes. Dave at col. 11, lines 30-34.
Datta in view of Jain and Dave does not expressly disclose the step of calculate a score of a product page indicating search probability and a coincidence between the first query and the first data item index. Here, a “first data item index” is an attribute of a data item (i.e., a product page). Datta does disclose receiving a user query to identify one or more query results comprising products (Datta at paras. 0006, 0014) based on machine learning to return more relevant products. Datta at para. 0017. In other words, Datta discloses a user performing a shopping query of a product catalog and returning data items (i.e., product pages) that satisfy the shopping query. Further, Datta discloses calculating a similarity of user query terms (i.e., first query) and metadata associated with the user query terms. Datta at para. 0076. However, Datta in view of Jain does not expressly disclose calculating a score of a product page (i.e., data item) indicating search probability and coincidence between the first query and the first data item index.
Aggarwal discloses a system and method for searching product catalogs by shoppers. Aggarwal at abstract. The system and method provides the user with the ability to search an online product catalog by performing similarity searches. The similarity search is performed by using a similarity function to calculate a similarity score for an object (i.e., a product page/data item) in the product database and the user query (i.e., calculate a score … between the first query and the first data item index). Aggarwal at paras. 0041-42.
Datta, Jain, Dave, and Aggarwal are analogous art because they are all directed to the same field of endeavor of searching for products.
At the time before the effective filing date of the instant application, it would have been obvious to one of ordinary skill in the art to modify Datta in view of Jain and Dave by adding the step of calculate a score of a product page indicating search probability and a coincidence between the first query and the first data item index, as disclosed by Aggarwal.
The motivation for doing so would have been to improve a user’s shopping experience and allows searches to be done on both numeric and nominal product attributes. Aggarwal at para. 0038.
In regards to claim 5, Datta in view of Jain, Dave, and Aggwaral discloses the search system according to claim 1, wherein the at least one processor is configured to train the learning machine to learn the relationship whenever the user selects a data item. Datta at paras. 0058-59, 0085-86.14
In regards to claim 6, Datta in view of Jain, Dave, and Aggwaral discloses the search system according to claim 1, wherein the at least one processor is configured to generate the learning machine based on a recurrent neural network model, a long short-term memory model, or a sequence conversion model. Datta at para. 0049.
In regards to claim 7, Datta in view of Jain, Dave, and Aggwaral discloses the search system according to claim 1, wherein the at least one processor is configured to:
a. add a new data item to the database (Datta at paras. 0044, 0057)15,
b. enter a first index of the new data item into the learning machine, and obtain a third query of the new data item from the learning machine (Datta at para. 0049, 0056-58)16, and
c. register the third query obtained from the learning machine in the database in association with the new data item. Datta at paras. 0070.17
In regards to claim 8, Datta in view of Jain, Dave, and Aggwaral discloses the search system according to claim 1, wherein
a. the database stores, for each data item, a first index and a second index that is registered (Datta at paras. 0057, 0060-70)18.
In regards to claim 10, Datta discloses a search method comprising:
a. searching a database, in which a plurality of data items are stored, using a first query entered by a user (Datta at para. 0073)19;
b. providing the user with a data item selected by the user from a search result (Datta at paras. 0042, 0084-85)20;
c. training a learning machine to learn a relationship between a first data item index of the data item as an input and the first query as an output (Datta at paras. 0057-60, 0063-64, 0070)21;
d. wherein the first data item index is an attribute of the data item (Datta at para. 0058)22;
e. entering a first target data item index of a target data item that is a new data item to the database into the learning machine and obtaining a second query that is output from the learning machine (Datta at paras. 0057-59, 0063-65)23; and
f. registering the second query as a second target data item index of the target data item (Datta at para. 0070)24;
g. wherein the second target data item index is not a category of the target data item (Datta at para. 0049)25; and
h. wherein the first query and the second query are inaccurate (Datta at para. 0016)26
i. record a combination of the first data item index of the data item selected by the user and a first character information used as the query (Datta at para. 0058)27, and
j. train the learning machine to learn the relationship based on the combination once a predetermined time and date have arrived (Datta at paras. 0049, 0085)28,
k. when the user inputs a new query, provide the user with a first search result based on the first data item index, the second target data item index, and the new query, (Datta at para. 0017)
l. when the user inputs the new query, provide the user with a second search result based on the second target data item index, the first data item index, and the new query (Datta at para. 0017)29
m. wherein the second target data item index is unique when compared to the first data item index (Datta at paras. 0059, 0063)30;
n. wherein the second target data item index does not contain items of the first data item index (Datta at paras. 0059, 0063)31;
Datta does not expressly disclose wherein the first search result and the second search result are provided on a same screen at a same time. As noted above, Datta does disclose providing a first search result and a second search result in response to a single query. However, Datta does not expressly disclose these results are both provided on a same screen at a same time.
Jain discloses a system and method for providing offers/coupons (i.e., second search result) within search results (i.e., first search result) in response to a query. Jain at paras. 0018-19. The offers/coupons can be determined based on keywords found in the query and user search behavior and intent. Jain at paras. 0042-43. These offers/coupons (i.e., second search result) are displayed along with the regular search results (i.e., first search result) on the same screen at the same time, in response to a single search query. Jain at Fig. 3.
Datta and Jain are analogous art because they are both directed to the same field of endeavor of searching for products of services and including additional related information.
At the time before the effective filing date of the instant Application, it would have been obvious to one of ordinary skill in the art to modify Datta by adding the feature of wherein the first search result and the second search result are provided on a same screen at a same time, as disclosed by Jain.
The motivation to do so would have been because it provides an efficient manner of present to the user relevant coupons/offers without the user providing additional input or performing additional steps. Jain at para. 0019. As discussed above, Datta already discloses providing two search results. Datta also discloses providing results that include advertisements and offers, but does not expressly disclose the results and offers are displayed on the same screen at the same time. Thus, one of ordinary skill in the art, relying on Jain, would have been motivated to modify Datta to display offers and coupons with the search results on the same screen at the same time for the benefit of efficiently providing the user offers without additional steps.
Datta in view of Jain does not expressly disclose wherein the character information entered by the user in the past for a product information is registered as the first data item index.
Dave discloses a method of generating attributes for an item (i.e., registering as the first data item index) utilizing customer queries (i.e., character information entered by the user in the past for a product information). Dave at col. 8, lines 22-36.
Datta, Jain, and Dave are analogous art because they are directed to the same field of endeavor of searching for products and services.
At the time before the effective filing date of the instant Application, it would have been obvious to one of ordinary skill in the art to modify Datta in view of Jain by adding the feature of wherein the character information entered by the user in the past for a product information is registered as the first data item index, as disclosed by Dave.
The motivation for doing so would have been to reduce and eliminate the need for merchants to complete an item summary that identifies attributes. Dave at col. 11, lines 30-34.
Datta in view of Jain and Dave does not expressly disclose the step of calculate a score of a product page indicating search probability and a coincidence between the first query and the first data item index. Here, a “first data item index” is an attribute of a data item (i.e., a product page). Datta does disclose receiving a user query to identify one or more query results comprising products (Datta at paras. 0006, 0014) based on machine learning to return more relevant products. Datta at para. 0017. In other words, Datta discloses a user performing a shopping query of a product catalog and returning data items (i.e., product pages) that satisfy the shopping query. Further, Datta discloses calculating a similarity of user query terms (i.e., first query) and metadata associated with the user query terms. Datta at para. 0076. However, Datta in view of Jain does not expressly disclose calculating a score of a product page (i.e., data item) indicating search probability and coincidence between the first query and the first data item index.
Aggarwal discloses a system and method for searching product catalogs by shoppers. Aggarwal at abstract. The system and method provides the user with the ability to search an online product catalog by performing similarity searches. The similarity search is performed by using a similarity function to calculate a similarity score for an object (i.e., a product page/data item) in the product database and the user query (i.e., calculate a score … between the first query and the first data item index). Aggarwal at paras. 0041-42.
Datta, Jain, Dave, and Aggarwal are analogous art because they are all directed to the same field of endeavor of searching for products.
At the time before the effective filing date of the instant application, it would have been obvious to one of ordinary skill in the art to modify Datta in view of Jain and Dave by adding the step of calculate a score of a product page indicating search probability and a coincidence between the first query and the first data item index, as disclosed by Aggarwal.
The motivation for doing so would have been to improve a user’s shopping experience and allows searches to be done on both numeric and nominal product attributes. Aggarwal at para. 0038.
Claim 11 is essentially the same as claim 10 in the form of a non-transitory computer-readable information storage medium for storing a program (Datta at para. 0052). Therefore, it is rejected for the same reasons.
In regards to claim 12, Datta in view of Jain, Dave, and Aggwaral discloses the search system according to claim 1, wherein the second query is generated by the learning machine. Datta at para. 0063-64, 0070.32
In regards to claim 14, Datta in view of Jain, Dave, and Aggwaral discloses the search system according to claim 1, wherein the second target data item index comprises feature information. Datta at para. 0066.
In regards to claim 17, Datta in view of Jain, Dave, and Aggwaral discloses the search system according to claim 1, wherein the second query is used to search the database. Datta at paras. 0077, 0079.33
In regards to claim 19, Datta in view of Jain, Dave, and Aggwaral discloses the search system according to claim 1, wherein the at least one processor configured to:
a. obtain an updated second query after a predetermined amount of time (Datta at paras. 0049, 0061-63, 0070);
b. register the updated second query in the database as the second target data item index of the target data item. Datta at paras. 0049, 0061-63, 0070.34
Claims 9 and 23 are rejected under 35 U.S.C. 103 as being unpatentable over Datta et al. (US Patent Pub 2018/0113919) (Datta), in view of Jain et al. (US Patent Pub 2014/0280339) (Jain), Dave et al. (US Patent 8,156,073) (Dave), and Aggarwal et al. (US Patent 2002/0138481) (Aggarwal), further in view of Zheng (US Patent Pub 2005/0267809).
In regards to claim 9, Datta in view of Jain, Dave, and Aggarwal discloses the search system according to claim 1, wherein
a. the data item relates to a product or service (Datta at para. 0014)35, and
b. the at least one processor is configured to provide the user with the first search result for displaying a page of the product or the service based on the first data item index (Datta at para. 0041, 0058, 0080)36, and provide the user with the second search result for providing a coupon of the product or the service. Datta at para. 0041.37 Jain at Fig. 3; para. 0043.
Datta in view of Jain, Dave, and Aggwaral does not expressly disclose the second search result is for providing a coupon of the product or the service is based on the second target data item index. As noted above, Datta does disclose presenting offers and advertisements to the user (i.e., second search result for providing a coupon) based on queries of the user. Datta at para. 0041.
Zheng discloses a system and method for presenting alerts to the user in a universal market system that aids a user in shopping. Zheng at para. 0037. Zheng uses information about a user’s shopping intent to determine features of the products the user wants to purchase. Using the information, it determines promotions/coupons that are currently available and presents them to the user. Zheng at Fig. 11; para. 0104.
Datta, Jain, Dave, Aggarwal, and Zheng are analogous art because they are both directed to the same field of endeavor of aiding a user to shop by learning user preferences and behavior.
At the time before the effective filing date of the instant application, it would have been obvious to one of ordinary skill in the art to modify Datta in view of Jain, Dave, and Aggwaral by adding the feature of the second search result is for providing a coupon of the product or the service is based on the second target data item index, as disclosed by Zheng.
The motivation for doing so would have been to provide users with information they may not have been aware of and that may last for a short duration. Zheng at para. 0104. As discussed above, Datta discloses providing the user with offers and advertisements related to their product query. Similarly, Zheng provides advertisements as well, but also searches for promotions for items the user has searched for.
In regards to claim 23, Datta in view of Jain, Dave, and Aggwaral discloses the search system according to claim 1, wherein the at least one processor is configured to:
a. obtain a coupon information of a product or a service (Jain at para. 0036);
b. send a display data of the second search result to a user terminal of the user (Jain at Fig. 3);
c. send a request for acquiring a coupon based on the coupon information and a selection of the coupon information by the user (Jain at para. 0042)38;
d. execute processing for enabling the user to acquire the coupon (Jain at para. 0042)39;
e. receive a registration request from a shop terminal including the product information (Datta at paras. 0014, 0044, 0047, 0056); and
f. add the product page to a product database based on the registration request. Datta at paras. 0014, 0044, 0047, 0056.
Datta in view of Jain, Dave, and Aggarwal does not expressly disclose the coupon information is based on the second target data item index, the first data item index and the new query as the second search result. It is noted in the rejection of claim 1, that Datta discloses performing a search on the database, which includes all the metadata (i.e., first data item index and second target data item index). What is not expressly disclosed is that the coupon information is retrieved based on product metadata/attributes.
Zheng discloses a system and method for presenting alerts to the user in a universal market system that aids a user in shopping. Zheng at para. 0037. Zheng uses information about a user’s shopping intent to determine features of the products the user wants to purchase. Using the information, it determines promotions/coupons that are currently available and presents them to the user (i.e., based on product metadata attributes). Zheng at Fig. 11; para. 0104.
Datta, Jain, Dave, Aggarwal, and Zheng are analogous art because they are both directed to the same field of endeavor of aiding a user to shop by learning user preferences and behavior.
At the time before the effective filing date of the instant application, it would have been obvious to one of ordinary skill in the art to modify Datta in view of Jain, Dave, and Aggwaral by adding the feature of the coupon information is based on the second target data item index, the first data item index and the new query as the second search result, as disclosed by Zheng.
The motivation for doing so would have been to provide users with information they may not have been aware of and that may last for a short duration. Zheng at para. 0104. As discussed above, Datta discloses providing the user with offers and advertisements related to their product query. Similarly, Zheng provides advertisements as well, but also searches for promotions for items the user has searched for.
Claims 13, 21, and 22 are rejected under 35 U.S.C. 103 as being unpatentable over Datta et al. (US Patent Pub 2018/0113919) (Datta), in view of in view of Jain et al. (US Patent Pub 2014/0280339) (Jain), Dave et al. (US Patent 8,156,073) (Dave), and Aggarwal et al. (US Patent 2002/0138481) (Aggarwal), further in view of Tavernier (US Patent 10,706,450).
In regards to claim 13, Datta in view of Jain, Dave, and Aggarwal discloses the search system according to claim 1, but does not expressly disclose wherein the first data item index comprises product titles. Datta does disclose storing metadata attributes about products. Datta at para. 0060.
Tavernier discloses product pages include an item title (i.e., product title). Tavernier at col. 3, lines 62-66.
Datta, Jain, Dave, Aggarwal, and Tavernier are analogous art because they are both directed to the same field of endeavor of aiding a user to shop by learning user preferences and behavior.
At the time before the effective filing date of the instant application, it would have been obvious to one of ordinary skill in the art to modify Datta in view of Jain, Dave, and Aggarwal by making the first data item index comprise product titles, as disclosed by Tavernier.
The motivation for doing so would have been because Datta in view of Jain, Dave, and Aggwaral already discloses storing metadata attributes of products, such as features, brands, categories, etc. but simply does not disclose an attribute as being a product title. A product title would simply be an additional attribute and would help identify products for selection, such as disclosed by Tavernier.
In regards to claim 21, Datta in view of Jain, Dave, and Aggwaral discloses the search system according to claim 1, wherein the second target data item index is a feature information. Datta at paras. 0065-69)40 but does not expressly disclose the first data item index is a product title.
Tavernier discloses product pages include an item title (i.e., product title). Tavernier at col. 3, lines 62-66.
At the time before the effective filing date of the instant application, it would have been obvious to one of ordinary skill in the art to modify Datta in view of Jain, Dave, and Aggarwal by making the first data item index a product title, as disclosed by Tavernier.
The motivation for doing so would have been because Datta in view of Jain, Dave, and Aggwaral already discloses storing metadata attributes of products, such as features, brands, categories, etc. but simply does not disclose an attribute as being a product title. A product title would simply be an additional attribute and would help identify products for selection, such as disclosed by Tavernier.
In regards to claim 22, Datta in view of Jain, Dave, and Aggarwal discloses the search system according to claim 1, but does not expressly disclose wherein when the user selects a new data item from the first search result or the second search result, the at least one processor is configured to display a product page including a link to a second product page for a similar product which is similar to the new data item. As noted in the rejection of claim 1 above, Datta does discloses presenting a user with search results and the user selecting one or more of them (i.e., viewing a product page).
Tavernier discloses a product page that includes the product and links to similar products to the product, which will send the user to the product page of the selected similar product. Tavernier at Fig. 1C; col. 4, lines 35-65; col. 5, lines 10-14.
At the time before the effective filing date of the instant application, it would have been obvious to one of ordinary skill in the art to modify Datta in view of Jain, Dave, and Aggarwal by adding the features of wherein when the user selects a new data item from the first search result or the second search result, the at least one processor is configured to display a product page including a link to a second product page for a similar product which is similar to the new data item, as disclosed by Tavernier.
The motivation for doing so would have been to show the user the recommendations are similar to the product they are viewing and allow the user to navigate products to achieve their goal of finding the product they want. Tavernier at col. 4, lines 54-58; col. 5, lines 11-21.
Claim 18 is rejected under 35 U.S.C. 103 as being unpatentable over Datta et al. (US Patent Pub 2018/0113919) (Datta), in view of in view of Jain et al. (US Patent Pub 2014/0280339) (Jain), Dave et al. (US Patent 8,156,073) (Dave), and Aggarwal et al. (US Patent 2002/0138481) (Aggarwal), further in view of Neal et al. (US Patent Pub 2003/0145277) (Neal).
In regards to claim 18, Datta in view of Jain, Dave, and Aggwaral discloses the search system according to claim 1, wherein the first query and the second query are inaccurate but does not expressly disclose the are inaccurate if they do not comprise a word included in the title of the target data item.
Neal discloses a system and method for searching an electronic catalog. Neal discloses the search can be broad and based on any characteristic of an item (i.e., feature information or attribute). The search string can include a part number of an item or any descriptive attribute of the item. The search engine can also handle misspellings, word fragments, or any other string that can lead a user to find a desired product (i.e., an inaccurate query). Neal at para. 0037. Since Neal discloses a search string (i.e., query) can include any attribute/characteristic of an item, such as a part number (i.e., does not comprise a word included in the title of the target data item), which is different from a name/title of an item, the search queries are interpreted as inaccurate queries.
Datta, Jain, Dave, Aggarwal, and Neal are analogous art because they are both directed to the same field of endeavor of searching products.
At the time before the effective filing date of the instant application, it would have been obvious to one of ordinary skill in the art to modify Datta in view of Jain, Dave, and Aggwaral by adding the features of the first query and the second query are inaccurate if they do not comprise a word included in the title of the target data item, as disclosed by Neal.
The motivation for doing so would have been to provide a search system that is able to provide the user with desired results while not requiring the user to enter an exact or specific search. Neal at para. 0037.
Claim 20 is rejected under 35 U.S.C. 103 as being unpatentable over Datta et al. (US Patent Pub 2018/0113919) (Datta), in view of Jain et al. (US Patent Pub 2014/0280339) (Jain), Dave et al. (US Patent 8,156,073) (Dave), and Aggarwal et al. (US Patent 2002/0138481) (Aggarwal), further in view of Applicant Admitted Prior Art (AAPA) (Specification at Fig. 10; pgs. 20, lines 23-27 to pg. 21, lines 1-16; pg. 25, lines 16-27 to pg. 26, lines 1-6; pg. 26, lines 26-27 to pg. 27, lines 1-22).
In regards to claim 20, Datta in view of Jain, Dave, and Aggwaral discloses the search system according to claim 1, wherein the at least one processor configured to implement a machine learning processor that uses a neural network model, such as a bidirectional recurrent neural network (Datta at para. 0046), but does not expressly disclose the specific steps of:
a. separate each of a plurality of words comprising the first target data index;
b. wherein the learning machine comprises a plurality of encoders and a plurality of decoders;
c. wherein each of the plurality of words comprising the first target data item index are entered into one of the plurality of encoders;
d. wherein the learning machine calculates and sequentially records a plurality of internal vectors indicating a plurality of internal states for the plurality of encoders based on the plurality of words;
e. wherein the learning machine obtains a first complete internal vector based on the plurality of internal vectors after all the plurality of words are entered into the plurality of encoders;
f. wherein the learning machine provides the first complete internal vector to one of the plurality of decoders and instructs the decoder to start an output;
g. wherein the one decoder outputs a first word and increments the first internal vector to a second complete internal vector;
h. wherein the second complete internal vector is input into the one decoder and a second word is output;
i. wherein the learning machine repeats incrementing a completed internal vector and outputting a new word until the one decoder reaches an end of outputs; and
j. wherein at least one of the words output by the one decoder is the second query.
As discussed in the rejection of claim 1, Datta in view of Jain, Dave, and Aggwaral discloses the “first target data index,” “the second query,” and the “learning machine.” What is not expressly disclosed are the specific steps performed by the learning machine when provided the “first target data index.”
AAPA describes the learning machine of the invention can utilizes recurrent neural networks as the machine learning algorithm that employs a sequence conversion model using long short term memory cell. Spec. at pg. 20, lines 23-27 to pg. 21, line 1-16. AAPA discloses Fig. 10 shows an example recurrent neural network. Fig. 10 shows the plurality of encoders and the plurality of decoders. The algorithm takes the product title (i.e., first target data index) and separates it out and inputs them into the encoders to receive internal vectors indicating internal states (i.e., context), which are subsequently input to the decoders (i.e., complete internal vector). Spec. at Fig. 10; pg. 25-26. Lastly, as admitted by Applicant, the machine learning algorithm is known in the art. Spec at pg. 27.
Datta, Jain, Dave, Aggarwal, and AAPA are analogous art because they are all directed to the same field of endeavor of searching for products.
At the time of the invention, it would have been obvious to one of ordinary skill in the art to modify Datta in view of Jain, Dave, and Aggwaral by adding the features of separate each of a plurality of words comprising the first target data index, wherein the learning machine comprises a plurality of encoders and a plurality of decoders, wherein each of the plurality of words comprising the first target data item index are entered into one of the plurality of encoders, wherein the learning machine calculates and sequentially records a plurality of internal vectors indicating a plurality of internal states for the plurality of encoders based on the plurality of words, wherein the learning machine obtains a first complete internal vector based on the plurality of internal vectors after all the plurality of words are entered into the plurality of encoders, wherein the learning machine provides the first complete internal vector to one of the plurality of decoders and instructs the one decoder to start an output, wherein the one decoder outputs a first word and increments the first internal vector to a second complete internal vector, wherein the second complete internal vector is input into the one decoder and a second word is output, wherein the learning machine repeats incrementing a completed internal vector and outputting a new word until the one decoder reaches an end of outputs, and wherein at least one of the words output by the one decoder is the second query, as disclosed by AAPA.
The motivation for doing so would have been because this type of machine learning algorithm is known in the art for processing data of undefined length.
Response to Arguments
Rejection of claims 1, 5-8, 10-12, 14, and 17 under 35 U.S.C. 103
Applicant’s arguments in regards to the rejections to claims 1, 5-8, 10-12, 14, and 17 under 35 U.S.C. 103, have been fully considered but they are not persuasive.
Applicant alleges Datta in view of Jain and Aggarwal fails to disclose (1) “wherein the second target data item index is unique when compared to the first data item index,” (2) “wherein the second target data item index does not contain items of the first data item index,” and (3) “wherein a character information entered by a past user for a product information is registered as the first data item index.”
Examiner is required to give claim limitations their broadest reasonable interpretation in light of the specification. However, limitations from the specification are not read into the claims. MPEP 2111.
In regards to limitations (2) and (3), Applicant responds to Examiner’s request for clarification and explains that the limitations are supported because “a first data item index” is based on a product title and the “second target data item index” is based on feature information. Remarks at 13. In other words, Applicant seems to contend that the “first data item index” is unique when compared to the “second target data item index” and the “second target data item index” does not contain items of the “first data item index” because they are attributes of different types (e.g., product title vs feature information). This interpretation of the limitations is still disclosed by Datta. Datta discloses that metadata attributes (i.e., first data item index) of a product can be any of a number of different types, such as a “category” or a “brand”. Similarly, inferred attributes can be any of these different types as well, including “feature tags” (i.e., feature information). Datta at paras. 0065-69. Accordingly, like the claimed invention, the metadata attributes (i.e., first data item index) and the inferred attributes (i.e., second target data item index) are unique when compared to each other and the “second target data item index” does not contain items from the “first data item index” because metadata attributes can be a ”category” or “brand” while the inferred attribute can be a “feature tag” (i.e., they are attributes of different types).
In regards to limitations (3) and (4), Applicant argues Datta does not disclose the limitations because the invention uses different ways that the first and second searches may be conducted. Remarks at 14-15. Examiner respectfully disagrees. The limitations simply require that a single query is performed on the database, which includes the “first data item index” and the “second target data item index” and returning two sets of search results based on both indexes and the user’s query. As disclosed by Datta, when a user enters a search query (i.e,. new query), the user is provided with the item previously selected by a user (i.e., first search result) and items associated with the previously selected item that have associated metadata (i.e., second search result). Datta at para. 0017. This meets the limitation because the query is executed against the database, which has metadata for all product items including the metadata for the selected item (i.e., first data item index) and metadata for associated items (i.e., second target data item index). Therefore, the two sets of search results are based on the user’s new query, the metadata of their previously selected item (i.e., first data item index) and the metadata of items associated with the previously selected item (i.e., second target data item index).
In regards to new claim 21, as explained above, Datta discloses metadata can be of different types, including a category or brand, as well as feature information. Accordingly, metadata of a selected data item can be a product name (i.e., product title) and inferred attributes can be “feature tags” (i.e., feature information).
In regards to new claim 22, the limitations are disclosed by previously cited prior art reference Tavernier. Tavernier discloses a product page with links to recommended products similar to the product on the product page. Tavernier at Fig. 1C.
In regards to new claim 23, the previously cited prior art discloses the limitations relating the coupon as set forth in the rejection above. The combination of cited art discloses obtaining coupon information based on product information and user query/intent (i.e., first data item index, second target data item index, and new query). As explained above, Datta discloses generating search results based on these elements. Datta also discloses providing offers. Datta at para. 0041. Zheng discloses obtaining coupon information for a product based on product information, which are the attributes of products. Zheng at para. 0104. In combination with Datta and Jain, Zheng discloses obtaining the coupon information based on the first data item index, the second target data item index, and the new query as the second search result. As further disclosed by Jain, the offer is presented to the user as a second search result on the same screen as the first search result. Jain at Fig. 3. Both Jain and Zheng discloses sending a request to acquire the coupon based on selection of the coupon information by the user and executing the processing to enable the user to acquire the coupon. Jain at para. 0042. Zheng at para. 0092. Lastly, Datta discloses the ability for a merchant to register and add new products, which add a product page to the database. Datta at paras. 0014, 0044, 0047, 0056.
Applicant does not present arguments in regards to the remaining limitations. Based on limitation (1), new grounds of rejection are set forth above as necessitated by Applicant’s amendments. The new grounds of rejection rely on Dave, which discloses an item attribute generation method and system that utilizes customer queries (i.e., character information entered by the user in the past for a product information …).
Rejection of claims 9, 13, and 18-20 under 35 U.S.C. 103
Applicant does not present arguments in regards to the rejections to claims 9, 13, and 18-20 under 35 U.S.C. 103. Consequently, the rejection to claims 9, 13, and 18-20 under 35 U.S.C. 103 remain rejected under the new grounds of rejection as necessitated by Applicant’s amendments.
Additional Prior Art
Additional relevant prior art are listed on the attached PTO-892 form. Some examples are:
Mukherjee et al. (US Patent 8,612,306) discloses a system and method for recommending products using category attributes.
Subramanya et al. (US Patent 11,182,840) discloses a system and method for mapping a predicted entity to a product based on a query.
Speers et al. (US Patent 7,725,363) discloses a system and method for generating product categories from a metadata tag.
Nakaji et al. (US Patent Pub 2018/0060923) discloses a system and method for desired advertisements and offers are displayed.
Conclusion
Applicant's amendment necessitated the new ground(s) of rejection presented in this Office action. Accordingly, THIS ACTION IS MADE FINAL. See MPEP § 706.07(a). Applicant is reminded of the extension of time policy as set forth in 37 CFR 1.136(a).
A shortened statutory period for reply to this final action is set to expire THREE MONTHS from the mailing date of this action. In the event a first reply is filed within TWO MONTHS of the mailing date of this final action and the advisory action is not mailed until after the end of the THREE-MONTH shortened statutory period, then the shortened statutory period will expire on the date the advisory action is mailed, and any extension fee pursuant to 37 CFR 1.136(a) will be calculated from the mailing date of the advisory action. In no event, however, will the statutory period for reply expire later than SIX MONTHS from the date of this final action.
Any inquiry concerning this communication or earlier communications from the examiner should be directed to Examiner Michael Le whose telephone number is 571-272-7970 and fax number is 571-273-7970. The examiner can normally be reached Mon-Fri 9:30 AM – 6 PM.
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, Tony Mahmoudi can be reached on 571-272-4078. 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.
/MICHAEL LE/Examiner, Art Unit 2163
/TONY MAHMOUDI/Supervisory Patent Examiner, Art Unit 2163
1 User performs a search of the product database (i.e., plurality of data items…) using an input query (i.e., first query).
2 Search results are returned and a user makes a selection, which provides the user with a product display (i.e., data item selected)
3 The machine learning algorithm is trained with the user input query (i.e., first query) and metadata attributes of the product selected by the user (i.e., first index of the data item). Based on the training, the machine learning algorithm is able to infer attributes (i.e., a query as output) based on attributes of a product (i.e., first index of the data item).
4 Metadata of a product is an attribute of the product (i.e., attribute of the data item).
5 The query system takes in new product information including metadata (i.e., first index of a target data item) and enters the new data into the machine learning algorithm, which is able to analyze the information based on stored relationships and infer attributes (i.e., second query output).
6 The inferred attributes are stored and used for future shopping queries.
7 Implied/inferred attributes can be any of the product metadata attributes, such as a feature tag.
8 The user’s query (i.e., first query) and the latent intent, which includes query terms (i.e., second query) are both inaccurate because the system must use the learned information to create a structured query based on previously submitted queries (like the first query) and determined latent intent (i.e., second query).
9 User query and metadata of a selected product are stored.
10 Updating the machine learning system (i.e., training) can be done at various times, such as when a user selects a result, makes a purchase, etc. (i.e., predetermined time and date). Alternatively, the BRRN can be updated periodically (i.e., predetermined time and date).
11 Datta discloses the next time a user inputs a query (i.e., a new query), the system returns not just the result of a previously selected product to the user (i.e., first search results based on the first index and the new query), but also other products associated with the catalog field/metadata terms associated with the selected product (i.e., second search results based on the second index and the new query).
12 The attributes identified based on the user query is interpreted as the first index. The inferred attributes of a product are interpreted as a second index. Note that a product attribute can also be a query. For example, a feature “red” can also be a query of a single word “red”.
13 The first index are the attributes/metadata identified with reference to the input user query, while the second index are inferred attributes based on analysis of the metadata and stored information with regards to user queries and actions. Therefore, the second index is interpreted as not containing items of the first index because there would be no need to infer the attributes if they already existed as the first index.
14 The machine learning algorithm is trained by the query system to learn the relationships based on user selections.
15 Merchants add product information to the database
16 The query system identifies metadata (i.e., first index) of the product and enters it into the machine learning algorithm. The machine learning system outputs inferred attributes and a query based on them (i.e., third query), which is stored for future use.
17 The inferred attributes are stored for future use with future shopping queries.
18 Products (i.e., data items) are stored with their metadata attributes, both identified (i.e., first index) and inferred (i.e., second index).
19 User performs a search of the product database (i.e., plurality of data items…) using an input query (i.e., first query).
20 Search results are returned and a user makes a selection, which provides the user with a product display (i.e., data item selected)
21 The machine learning algorithm is trained with the user input query (i.e., first query) and metadata attributes of the product selected by the user (i.e., first index of the data item). Based on the training, the machine learning algorithm is able to infer attributes (i.e., a query as output) based on attributes of a product (i.e., first index of the data item).
22 Metadata of a product is an attribute of the product (i.e., attribute of the data item).
23 The query system takes in new product information including metadata (i.e., first index of a target data item) and enters the new data into the machine learning algorithm, which is able to analyze the information based on stored relationships and infer attributes (i.e., second query output).
24 The inferred attributes are stored and used for future shopping queries.
25 Implied/inferred attributes can be any of the product metadata attributes, such as a feature tag.
26 The user’s query (i.e., first query) and the latent intent, which includes query terms (i.e., second query) are both inaccurate because the system must use the learned information to create a structured query based on previously submitted queries (like the first query) and determined latent intent (i.e., second query).
27 User query and metadata of a selected product are stored.
28 Updating the machine learning system (i.e., training) can be done at various times, such as when a user selects a result, makes a purchase, etc. (i.e., predetermined time and date). Alternatively, the BRRN can be updated periodically (i.e., predetermined time and date).
29 Datta discloses the next time a user inputs a query (i.e., a new query), the system returns not just the result of a previously selected product to the user (i.e., first search results based on the first index and the new query), but also other products associated with the catalog field/metadata terms associated with the selected product (i.e., second search results based on the second index and the new query).
30 The attributes identified based on the user query is interpreted as the first index. The inferred attributes of a product are interpreted as a second index. Note that a product attribute can also be a query. For example, a feature “red” can also be a query of a single word “red”.
31 As explained in FN24, the first index are the attributes/metadata identified with reference to the input user query, while the second index are inferred attributes based on analysis of the metadata and stored information with regards to user queries and actions. Therefore, the second index is interpreted as not containing items of the first index because there would be no need to infer the attributes if they already existed as the first index.
32 The machine learning model infers (i.e., generates) attributes to be used for future shopping queries (i.e., second query). The inferred attributes are interpreted as the second query
33 Inferred attributes (i.e., second query) is used to assemble a query to produce predictable, reasonable query results.
34 The learning machine continually or periodically (i.e., after a predetermined amount of time) updates itself with new information, which includes metadata/attributes (i.e., second data item query) that are used to infer attributes based on received queries.
35 The system is used for a product searching/shopping.
36 A product display (i.e., product page) is provided to the user in response to a user search, which is associated with product attributes (i.e., first index) used to retrieve the product.
37 Offers, advertisements (i.e., a coupon of the product or service) is provided to the user.
38 Upon selection, the coupon information is retrieved to send to the user.
39 The coupon can be sent to the user in some way.
40 Attributes can be a category or brand or they can be feature tags (i.e., feature information).