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 .
This communication is in response to the Amendment filed 07/30/2025.
Response to Arguments
Claims 1 – 28 are pending in this Office Action. After a further search and a thorough examination of the present application, claims 1 – 28 remain rejected.
Applicant's arguments filed with respect to claims 1 – 28 have been fully considered but they are moot in view of new rejection.
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 pre-AIA 35 U.S.C. 103(a) which forms the basis for all obviousness rejections set forth in this Office action:
(a) A patent may not be obtained though the invention is not identically disclosed or described as set forth in section 102, if the differences between the subject matter sought to be patented and the prior art are such that the subject matter as a whole would have been obvious at the time the invention was made to a person having ordinary skill in the art to which said subject matter pertains. Patentability shall not be negated by the manner in which the invention was made.
The factual inquiries for establishing a background for determining obviousness under pre-AIA 35 U.S.C. 103(a) 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 – 28 are rejected under pre-AIA 35 U.S.C. 103(a) as being unpatentable over Bender et al. (US 2023/0289868 A1) (‘Bender’ herein after) further in view of Chao et al. (US 2024/0403373 A1) (‘Chao’ herein after) further in view of Hughes et al. (US 2017/0357717 A1) (‘Hughes’ herein after) further in view of Wang et al. (US 2025/0307318 A1) (‘Wang’ herein after).
With respect to claims 1, 8, 15,
Bender discloses a computer-implemented method performed by one or more processors, the computer-implemented method comprising: receiving a request from a client device to return a set of content items (figures 5, 6, paragraph 86 and 92 teach a search bar where a query or user input may be entered, Bender); generating, by the one or more processors, a search criterion to search for content items responsive to the request (figures 5, 6, paragraphs 86 – 88 and 92 teach a search criteria and items/ categories matching in a database, Bender); generating, by the one or more processors and utilizing the search criterion a set of relevancy-ranked content items, by a generative AI search and retrieval system (figures 5, 6, paragraphs 49 – 50 teaches items and taxonomy that are relevant to the search, paragraphs 85 – 95, Bender); determining a plurality of utility scores corresponding toa plurality of display groupings where a content item could be assigned, assigning by performing an allocation process on the plurality of display groupings individually and in parallel, the content item of the set of relevant ranked items to a display grouping of the plurality of display groupings according to the plurality of utility score (figures 3 (#322), 5, 6 and paragraphs 67 – 88 teaches where it teaches various groupings and where multiple labels maybe obtained for an item and the labels are from different level of the taxonomy, to resolve this machine-learned carousel model identifying a category with the greatest relevance probability, Bender); generating, by the one or more processors, a carousel display of the set of relevancy-ranked content items, comprising a number of display groupings according to a predetermined number of display groupings, wherein each display grouping includes a predetermined number of display slots, representing an ordered position in a respective display grouping (figures 5, 6, paragraphs 52 – 56 teach the relevancy ranking of items, paragraphs 85 – 95, Bender); transmitting the carousel display of the relevancy-ranked content items and the content items to the client device and causing a user interface of the client device to display the set of relevancy-ranked content items according to the carousel display (figures 5, 6, paragraphs 65 teaches the transmitting and displaying of the carousels of the groups of results of the search data, paragraphs 85 – 95, Bender).
Bender teaches relevancy ranked items returned but does not specifically state the generative search and retrieval explicitly as claimed.
However, Chao teaches the generative search and retrieval in paragraphs 111 – 114 teaching the generative engine is utilized with the embedding engine and can thus expand dynamically the listing.
It would have been obvious to one of ordinary skill in the art before the effective filing date of the claimed invention because both references are in the same field of study, namely retrieval of relevant search data. Furthermore, Chao teaches techniques for ranking retrieved data listings based on a large language model (LLM). In some embodiments, a collection of data listings of a data exchange may be processed through an LLM to generate embeddings for each of the listings. The embeddings may represent a vector that describes the data listing within a logical space (e.g., a semantic and/or syntactic space). A search engine of the data exchange may receive (from a user) a search query comprising a set of search terms, and retrieve a set of data listings based on the search terms of the search query. The search query may also be processed by the LLM to generate an embedding for the search query. A data ranking module of the search engine may analyze the embeddings for the data listings returned by the search query as well as the embedding for the search query to determine which of the data listings are most relevant to the search query, and the data listings may be ranked based on the determined relevance to the search query, paragraphs 40 – 42, Chao.
Bender teaches the carousel display but does not teach the display structure definition explicitly as claimed.
However, Hughes teaches the display structure definition of the groupings in figures 4, 16 A-B and in paragraphs 114, 170, 176 – 177 teaching the positions in the templates of the display groupings and how the structure if this display is defined using the interface. Paragraphs 181 – 187 teaches the minimum and maximum of items/documents allowed in a group and the positioning with the branching factor with dynamic grouping with updated items.
It would have been obvious to one of ordinary skill in the art before the effective filing date of the claimed invention because both references are in the same field of study, namely retrieval of relevant search data. Furthermore, Hughes teaches that the templates used to position panes and section layouts is based on variety of variables including group size, sizes of items, titles, etc., thus the layout is generated in a manner that positions documents within the section to provide a visual effect with a descending level of visual density. Positioning documents with higher visual density at the top of each section provides a pleasing visual effect that draws a user's attention through the different sections of the aggregated feed, paragraphs 130 – 132, Hughes.
Bender teaches generating using the search criterion a set of relevancy-ranked content items, by a generative AI search and retrieval system, but does not teach generating descriptions for the content items and perform a vector similarity matching by comparing a search vector embedding generated from the search criterion to a set of vector embeddings generated from the descriptions for the content items explicitly as claimed.
However, Wang teaches generating descriptions for the content items and perform a vector similarity matching by comparing a search vector embedding generated from the search criterion to a set of vector embeddings generated from the descriptions for the content items in figures 2, 3, 6, 8 and paragraphs 50 – 52 teaching identifies data that is semantically similar to question. In examples, data identifier identifies data by matching search terms of query to keywords of data files (e.g., tag(s) of a data file, text within a data file, text within a description of a data file), by matching text of question to keywords of data files, by measuring similarity between embedding(s) that semantically describe question (also referred to as “question embeddings” herein) and embedding(s) that semantically describe data files (also referred to as “data embeddings” herein), and/or by performing another method to determine data is semantically similar to question, as described elsewhere herein. Furthermore, paragraph 59 – 62 teach question generator of FIG. 2 determines additional context based on a domain of search engine system that received query via user interaction, a keyword included in query, a filter applied to query, and/or any other information that question generator may analyze to determine additional context of query. Then a prompt to cause the LLM to generate the question is generated, the prompt comprising the first search term and the additional context.
It would have been obvious to one of ordinary skill in the art before the effective filing date of the claimed invention because both references are in the same field of study, namely retrieval of relevant search data. Furthermore, Wang teaches intelligently interpreting a search query and responding to the search query. For instance, in an example embodiment, a first search query comprising a first search term is received. A generative artificial intelligence model is used to generate a question based on the first search term. Data semantically similar to the question is identified. The generative AI model is used to determine an answer to the question based on the question and the identified data. A question-answer pair comprising the question and the answer is generated, paragraph 3, Wang.
With respect to claims 2, 9, 16,
Bender as modified discloses the computer-implemented method of claim 1, wherein generating the carousel display structure definition of the relevancy-ranked content items comprises: assigning a subset of the set of relevancy-ranked content items into the predetermined number of display groupings, wherein the predetermined number of display groupings includes 2 or more display groupings (figures 5, 6 and paragraphs 77 – 84 teach the relevancy and ranking of the items in groups and carousels, Bender).
With respect to claims 3, 10, 17,
Bender as modified discloses the computer-implemented method of claim 2, further comprising: determining a group order position for the display grouping of the predetermined number of display groupings by: determining a grouping score of the content items assigned to a set of a predetermined number of display slot positions for the display grouping wherein the grouping score is an aggregate relevancy value of the content items of the display grouping and sorting the display grouping according to the grouping score (figures 5, 6, paragraphs 67, 78 – 80 and 83, Bender).
With respect to claims 4, 11, 18,
Bender as modified discloses the computer-implemented method of claim 1, wherein generating the set of relevancy-ranked content items comprises: causing the search criterion to be processed, via the generative AI search and retrieval system, comprising one or more machine learning models; and generating by the generative AI search and retrieval system, the set of relevancy-ranked content items in response to the search criterion, wherein the set of relevancy-ranked content items have a content identifier and a content description (paragraphs 59 – 60 and 94, Bender).
With respect to claims 5, 12, 19,
Bender as modified discloses the computer-implemented method of claim 1, wherein generating the set of relevancy ranked content items comprises: performing, by the one or more processors, a first stage scoring process, to generate the set of content items (paragraph 97 states ”The embedding corresponding to the search query 502 may then be used to search for nearest neighbors to the embedding in the embedding store 516 from among the retrieved data listings 423, which may contain embeddings for each of the data listings 423 of the data exchange” the 1st stage output a subset of data listings 423 using nearest neighbor search, Chao) and performing, by the one or more processors, a second stage scoring process, by an inferencing machine learning model trained to determine item score values to generate a set of item score values for the set of content items (paragraph 97 “… the data listings 423 corresponding to the nearest-neighbor embeddings may be passed to a next phase where, for each retrieved listing, information from the corresponding embedding is combined with other signals to compute the final aggregated sum score for each data listing 423 of the retrieved data listings 423…” and paragraph 99 “The use of the LLM 510 to process and/or rank search results from a search query 502 may provide a number of benefits”, Chao).
With respect to claims 6, 13, 20,
Bender as modified discloses the computer-implemented method of claim 5, further comprising: performing, by the one or more processors, a third stage scoring process that modifies an item score value to adjust a content item position placement in the relevancy-ranked content items (paragraph 120 “the output of the LLM 510 may be the ordinal ranking of the data listing embeddings 710”, paragraph 121 “the output scores of the LLM 510 may not always be sufficient for getting a high-quality ranking. As a result, the nearest-neighbor results from the LLM 510 from block 604 may be combined with data listing signals” and paragraph 123 “Based on the data listing signals, the ranking order of the results returned by the LLM 510 may be adjusted…”, Chao).
With respect to claims 7, 14, 21,
Bender as modified discloses the computer-implemented method of claim 5, further comprising: performing, by the at least one processors, the allocation process that generates annotations for one or more of the content items in the relevancy-ranked content items, wherein the generated annotations comprise a type of content item for a respective content item in the relevancy-ranked content items (paragraph 131 “FIG. 9, the top data listings 423 of results may be examined and, for each one, the LLM 510 may be prompted to generate the listing explanation 916 explaining why the listing is relevant to the user's search query 502… The use of the listing explanation 916 may enable the user to make better decisions in adopting or discarding the data listings 423 of the results”, the listing explanations explain how relevant (i.e., type) of the data listing to the user query, Chao).
With respect to claims 22,
Bender as modified discloses the computer-implemented method of claim 1, further comprising: receiving a parameter, along with the request, from the client device setting the predetermined number of display groupings and the predetermined number of display slots for a display grouping (paragraph 77 – 80, Bender and paragraphs 170, 176 – 177, Hughes).
With respect to claims 23,
Bender as modified discloses the computer-implemented method of claim 1, further comprising: determining an item page size and a group page size of the client device and determining the predetermined number of display groupings and display slots to create based on the item page size and the group page size (paragraph 77 – 80, Bender and paragraphs 170, 176 – 177, Hughes).
With respect to claims 24,
Bender as modified discloses the computer-implemented method of claim 1, wherein a retrieved content item includes a preassigned display grouping value, and wherein the content item is to be assigned to a display grouping the preassigned display grouping value (figures 5, 6, paragraphs 67 and 83, Bender and figures 4, 16 A-B and in paragraphs 114, 170, 176 – 177, Hughes).
With respect to claims 25,
Bender as modified discloses the computer-implemented method of claim 1, further comprising: receiving from the client device an additional request for additional content items associated with a displaying grouping; retrieving additional content items that are similar to the content items of the display grouping and adding the additional content times to additional display slots for the particular display grouping (paragraphs 181 – 187, Hughes).
With respect to claims 26,
Bender as modified discloses the computer-implemented method of claim 1, further comprising: determining a display grouping position or a display slot position within the display grouping based on an associated attribute of the content item, wherein the associated attribute indicates an age of the content item (paragraphs 79 – 80, Bender and paragraphs 170 and 176 – 177, Hughes).
With respect to claims 27,
Bender as modified discloses the computer-implemented method of claim 1, further comprising: determining a plurality of display grouping relevancy score for the display groupings; based on a the plurality of display groupings relevancy score, determining that a particular display grouping is to be omitted from the display structure definition and assigning the content contents of the particular display grouping to be omitted to a plurality of other remaining display groupings (figures 5, 6, paragraphs 52 – 56, Bender and paragraphs 181 – 187, Hughes).
With respect to claims 28,
Bender as modified discloses the computer-implemented method of claim 1, further comprising: determining plurality of a relevancy scores for the plurality of display groupings, wherein the plurality of relevancy scores is based on content items assigned to the display grouping and based on the plurality of relevancy scores, changing an order of the plurality of display groupings (paragraphs 79 – 80, Bender and paragraphs 170 and 176 – 177, Hughes).
Prior Art
The prior art made of record and not relied upon is considered pertinent to applicant's disclosure.
US 20140101147 A1 teaches match one or more words in a received word query with one or more attribute, content and/or measure in or corresponding to the stored responses, to provide relevant responses to the input word query based on the matching, and to rank the relevant responses, wherein the ranking of each relevant response is based at least partially on the position of one or more matched attributes of that relevant response within at least one ordered combination.
US 20210334886 A1 teaches dynamically generating content that is organized and presented according to a strategy and/or a user context for a given shopping experience and allowing the user associated with the user account to rearrange the ordering of the displayed content.
US 10891676 B1 teaches a recommendation system groups the related items based on their respective attribute values and produces data that results in a user interface that displays the related items in these groups. The recommendation system generates labels for these groups such that a user can clearly identify what types of related items are included therein.
US 20210382952 A1 teaches and identifies allocations of both organic and promoted content on a given page. The allocations of page content are compared against one another and configured to prioritize for overall utility based on objective factors that quantify a page “look and feel” as measured by machine learning models.
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.
Contact Information
Any inquiry concerning this communication or earlier communications from the examiner should be directed to NAVNEET K GMAHL whose telephone number is 571-272-5636.
If attempts to reach the examiner by telephone are unsuccessful, the examiner’s supervisor, SANJIV SHAH can be reached on (571) 272-4098. The fax phone number for the organization where this application or proceeding is assigned is 571-273-8300.
Information regarding the status of an application may be obtained from the Patent Application Information Retrieval (PAIR) system. Status information for published applications may be obtained from either Private PAIR or Public PAIR. Status information for unpublished applications is available through Private PAIR only. For more information about the PAIR system, see http://pair-direct.uspto.gov. Should you have questions on access to the Private PAIR system, contact the Electronic Business Center (EBC) at 866-217-9197 (toll-free). If you would like assistance from a USPTO Customer Service Representative or access to the automated information system, call 800-786-9199 (IN USA OR CANADA) or 571-272-1000.
/NAVNEET GMAHL/Examiner, Art Unit 2166 Dated: 11/13/2025
/SANJIV SHAH/Supervisory Patent Examiner, Art Unit 2166