DETAILED ACTION
This action is in response to the application filed 3/4/2024 and subsequent amendment filed 7/1/2024.
Claims 1-20 have been submitted for examination.
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 .
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, 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.
Claim(s) 1-20 is/are rejected under 35 U.S.C. 103 as being unpatentable over Amoateng et al. (US 2024/0320257), in view of Bathwal et al. (US 2024/0281487), hereinafter Bathwal.
As per claim 1, Amoateng teaches the following:
an apparatus comprising:
a memory, (see Fig. 6, 650); and
a processor coupled to the memory, (see Fig. 6, 610), the processor configured to:
train an artificial intelligence (Al) model using a neural network capability to generate infused articles based on at least one of article content, user data, and model feedback data. As Amoateng teaches in paragraph [0033], a large language model is a large neural-network model with is trained on a large dataset. Amoateng teaches in paragraph [0032] that the large language model makes predictions about a user based upon user information that is known (user data) and teaches in paragraph [0046] that prompts are provided to the large language model including a document, i.e., an article (article content),
detect a request to open an article via a software application. As Amoateng teaches in paragraph [0074], and corresponding Fig. 4, at step 491 a request for ranked content (articles) is received,
retrieve data associated with a user of the software application from a data store. As Amoateng further teaches in paragraph [0074], in steps 493 and 494, a user profile is requested and received,
execute the trained Al model on the data associated with the user and the article to generate unique article content that includes a description that is related to the user and the article, and infuse the unique article content . As Amoateng teaches in paragraph [0075], a prompt that includes natural language text instructions is provided to generate pills. Amoateng teaches in paragraph [0023] that the feed items includes “a personalized title, a personalized summary of the corresponding content item, and a personalized picture”. Amoateng further teaches in paragraph [0036] that the personalization of the different elements is performed by the large language model when the model receives the user profile and content, and
open a screen of the software application and display the fused article of content within a graphical user interface (GUI) of the screen. See Fig. 3.
While Amoateng teaches of generating personalized pills, Amoateng does not explicitly teach of infusing this unique content in between existing content. In a similar field of endeavor, Bathwal teaches of a method of summarizing multiple documents (see abstract). Bathwal further shows in Fig. 3, and corresponding paragraph [0062], that citation controls may be inserted “in between” a generated answer (existing content).
It would have been obvious to one of ordinary skill in the art before the effective filing date of applicant’s claimed invention to have modified the pill controls of Amoateng to be placed in between content as in Bathwal. One of ordinary skill would have been motivated to have made such modification because the placement of Bathwal would benefit users in displaying the exact content to which the pill control relates.
Regarding claim 2, Amoateng teaches the apparatus of claim 1 as described above. Amoateng further teaches the following:
the processor is configured to execute the trained Al model on a plurality of articles of content to generate a new article of content, and infuse the unique article content into the new article of content. As Amoateng teaches in paragraph [0067], and corresponding Fig. 3, an example user feed is presented (new article) comprising summary information for a plurality of articles.
Regarding claim 3, Amoateng teaches the apparatus of claim 1 as described above. Amoateng further teaches the following:
the processor is further configured to receive a user identifier from the software application and query the data store for the user data based on the user identifier, wherein the data comprises one or more of social media data, a user profile, and account data. As Amoateng teaches in paragraph [0026], that content collected includes personal contacts in a social media network and the user profile is determined after the user logs on the service (user identifier).
Regarding claim 4, Amoateng teaches the apparatus of claim 1 as described above. However, Amoateng does not explicitly teach of utilizing a browsing history. Bathwal further teaches the following:
the processor is configured to ingest a browsing history from a browser installed on a source device associated with the request, identify contextual attributes of the user from the browsing, and further execute the trained Al model on the contextual attributes of the user to generate the fused article of content. As Bathwal teaches in paragraph [0041], a component of user information which may be stored and utilized is that of user search history.
It would have been obvious to one of ordinary skill in the art before the effective filing date of applicant’s claimed invention to have modified the user profile of Amoateng to include the user search history Bathwal. One of ordinary skill would have been motivated to have made such modification because Amoateng suggests utilizing similar information in paragraph [0062] that user interaction history and in paragraph [0026] that content collected may include content which the user has subscribed, such as news posts and services.
Regarding claim 5, Amoateng teaches the apparatus of claim 1 as described above. Amoateng further teaches the following:
the processor is further configured to execute a machine learning model on the data associated with the user to identify an objective of the user and further execute the trained Al model on the objective of the user identified by the machine learning model to generate the fused article. As Amoateng teaches in paragraph [0027], the user profile may include topics of interest, which the Examiner interprets as encompassing an “objective”.
Regarding claim 6, Amoateng teaches the apparatus of claim 1 as described above. Amoateng further teaches the following:
the processor is configured to generate a clickable link to a page associated with the unique article content based on execution of the trained Al model on data associated with the user and article and insert the clickable link into the fused article. As Amoateng teaches in paragraph [0036], each feed item may include a link to the corresponding item.
Regarding claim 7, Amoateng teaches the apparatus of claim 6 as described above. Amoateng further teaches the following:
the processor is configured to embed the clickable link into the unique article content within the fused article and activate the clickable link such that when clicked on, the graphical user interface navigates to a web page of content associated with the unique article content. As Amoateng teaches in paragraph [0036], each feed item includes a means for accessing the corresponding content item.
Regarding claim 8, Amoateng teaches the apparatus of claim 1 as described above. However, Amoateng does not explicitly teach of utilizing feedback to retrain the AI model. Bathwal further teaches the following:
the processor is further configured to receive feedback about the fused article via the graphical user interface, generate a model feedback record which includes the fused article and the feedback, add the model feedback record to the model feedback data, and retrain the Al model based on the model feedback data including the added model feedback record. As Bathwal teaches in paragraph [0028], a reward modeling component uses user feedback to continuously improve the quality of generated answers. Further see paragraph [0141] where feedback is utilized in “reinforcement learning”, i.e., retraining.
It would have been obvious to one of ordinary skill in the art before the effective filing date of applicant’s claimed invention to have modified the learning system of Amoateng to include the reinforced learning based on user feedback of Bathwal. One of ordinary skill would have been motivated to have made such modification because As Bathwal teaches in paragraph [0028], a reward modeling component uses user feedback to continuously improve the quality of generated answers.
As per claim 9, the limitations of claim 9 are substantially similar to those of claim 1 and are rejected using the same reasoning.
Regarding claims 10-16, modified Amoateng teaches the method of claim 9 as described above. The remaining limitations of claims 10-16 are substantially similar to those of claims 2-8 respectively, and are rejected using the same reasoning.
As per claim 17, Amoateng teaches the following:
a computer-readable storage medium comprising instructions stored therein which when executed by a processor. See Fig. 6.
The remaining limitations of claim 17 are substantially similar to those of claim 1 and are rejected using the same reasoning.
Regarding claim 18-20, modified Amoateng teaches the medium of claim 17 as described above. The remaining limitations of claim 18-20 are substantially similar to those of claims 2, 3, and 8 respectively, and are rejected using the same reasoning.
Conclusion
The prior art made of record and not relied upon is considered pertinent to applicant's disclosure.
-Wexler (US 2015/0193540), generating personalized news feed.
-Lyren (US 2014/0359439), see paragraph [0114].
Any inquiry concerning this communication or earlier communications from the examiner should be directed to GREGORY A DISTEFANO whose telephone number is (571)270-1644. The examiner can normally be reached Monday - Friday: 9 am - 5 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, William Bashore can be reached at 5712424088. 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.
/GREGORY A. DISTEFANO/
Examiner
Art Unit 2174
/WILLIAM L BASHORE/ Supervisory Patent Examiner, Art Unit 2174
Prosecution Insights