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 .
Claims 1, 6-9, 11-12, 14, & 19-20 have been amended.
Claims 1-20 are pending.
Response to Arguments
Applicant’s arguments with respect to claim(s) 1-20 have been considered but are moot because the new grounds of rejection. See Office Action below.
Claim Rejections - 35 USC § 102
The following is a quotation of the appropriate paragraphs of 35 U.S.C. 102 that form the basis for the rejections under this section made in this Office action:
A person shall be entitled to a patent unless –
(a)(1) the claimed invention was patented, described in a printed publication, or in public use, on sale, or otherwise available to the public before the effective filing date of the claimed invention.
Claim(s) 1-3, 7, 11-16, & 20 is/are rejected under 35 U.S.C. 102(a)(1) as being unpatentable over SHIBUI (US Pub. No. 2022/0004916 A1).
In respect to Claim 1, SHIBUI teaches:
an electronic device comprising: memory storing one or more instructions; a communication circuit; and a processor operatively coupled to the memory and the communication circuit, wherein the one or more instructions, when executed by the processor, cause the electronic device to: identify a use pattern of specified media items from a plurality of media items stored in the memory, (SHIBUI teaches [0029, 0043, 0059] that the user may select one of a plurality of buttons with regard to an item displayed on the user device. The user may select a like button, an unlike button or a next button, and such activity may be related to an item among the one or more items in the first representation and/or the at least one modality of the item, which is a per-item use pattern signal.)
determine, based on the use pattern, a score of each of the specified media items, (SHIBUI teaches [0100-0101] the device determines the user interest values for image and the user interest values for text with regard to the items, which is a per item, per modality score derived directly from the logged use-pattern activity.)
extract, using a main AI model stored in the memory, a feature corresponding to a characteristic of each of the specified media items, (SHIBUI teaches [0028, 0039, 0057] the preprocessed item data may include an extracted feature vector for each modality of each item, which is a main-model feature extraction stage producing a per item characteristic vector.)
train a first personalized AI model based on the score and the feature, (SHIBUI teaches [0045, 0062] a user device may train the at least one first prediction model based on the user activity log data to generate the at least one second prediction model, wherein the second prediction model (the personalized model) is trained from the activity-log derived interest values [0028, 0039, 0057] together with the item feature vectors, wherein the prediction module may calculate a distance value using the feature vector to predict the one or more items based on the at least one first prediction model.)
based on a request to perform a function related to first media items from the plurality of media items, determine, using the first personalized AI model, a first preference of each of first media items, (SHIBUI teaches [0040] in response to the receiving of the input, user device may obtain the preprocessed item data from the cache module and predict the one or more items that are related with the input from the user, and [0050] predict, based on the at least one second prediction model, one or more items, wherein a received request triggers the personalized model’s preference and prediction step.)
and based on the first preference, perform the function related to the first media items (SHIBUI teaches [0050] that following prediction, the device may sort the one or more items, and filtering some or changing the position of the items in the displayed representation.)
As per Claim 2, SHIBUI teaches:
recommend at least one media item the first media items based on the first preference (SHIBUI teaches [0050] that the sorted/filtered/repositioned items are re-presented to the user as a second representation [Abstract] of the predicted one or more items, wherein the model’s preference determination directly drives which items are recommended to the user.)
As per Claim 3, SHIBUI teaches:
based on the first preference, classify the first media items (SHIBUI teaches [0050] that the device may filter some of the one or more items or change the position of items in the displayed representation based on the prediction which is a preference driven partitioning or grouping of items that functions as a classification.)
As per Claim 7, SHIBUI teaches:
based on the score, determine the specified media items from the plurality of media items stored in the memory as a media set for training the first personalized AI model (SHIBUI teaches [0045] if a second evaluation value is below a predetermined value, the user device may select the underperforming model and retrain it on the accumulated activity-log data. This is a score gated (evaluation value gated) trigger for retraining.)
As per Claim 11, SHIBUI teaches:
acquire information on at least one media item from an external electronic device or a server, and train, using the information on the at least one media item, the first personalized AI model (SHIBUI teaches [0121, 0129] information received from the external apparatus and used in training the personal model.)
As per Claim 12, SHIBUI teaches:
identify, using the first personalized AI model, an attribute of each of the first media items, and based on the attribute, classify the first media items (SHIBUI teaches [0057] that the personalized prediction model computes a distance value using the feature vector for each item, which [0050] then is used to sort, filter, or change the position of items, which is a classification like function performed on the basis of that attribute.)
As per Claim 13, SHIBUI teaches:
identify the use pattern based on at least one of viewing, sharing, editing, deleting, a favorites setting, or a background setting of the specified media items (SHIBUI [0043, 0082])
Claims 14-16 are the method claims corresponding to device claims 1-3 respectively, therefore are rejected for the same reasons noted previously.
Claim 20 is the media claim corresponding to device claim 1 above, therefore is rejected for the same reasons noted previously.
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) 4 & 17 is/are rejected under 35 U.S.C. 103 as being unpatentable over SHIBUI, and further in view of GERACI et al. (US Pub. No. 2022/0004874 A1) & Duddu et al. (US Patent No. 8,762,462 B1).
As per Claim 4, GERACI teaches:
based on determining a second AI model of another person is acquired from an external electronic device, (GERACI teaches [0034, 0055, 0121] that the communication unit may receive, from the external apparatus, group model data collected with respect to a plurality of users – an external device sourced model is obtained over a communication interface.)
It would have been obvious to one of ordinary skill in the art at the time of the filing date of the invention to incorporate the teachings of GERACI into the system of SHIBUI & Duddu. One of ordinary skill in the art would be motivated to provide a highly reliable trained personal model even when personal data for training the personal model is insufficient. (GERACI [0005])
Dudda teaches:
determine, using the second AI model, a second preference of each of the first media items; and based on the second preference, perform a function related to at least one media item among the first media items (Dudda teaches [column 2, lines 35-55] an interest profile is a collection of data associated with a user, contact of a user, or group of users reflecting the types of content that contact favorable acts on a second other person specific model distinct from the user’s own. Dudda teaches [column 3, lines 24-34] at least one contact is identified as a suggested recipient of the unpublished post of the user based on the interest profile of the contact and the type of content of the unpublished post of the user, wherein the contact’s own interest profile model is run directly against the same content the user already has, producing a second, contact specific preference determination. Dudda teaches [column 6, lines 9-35] a suggest recipient can be determined if the respective contact’s interest profile has a ratio of interest for the type of content that is above a threshold, wherein the second model derived value gates a downstream function. Dudda teaches [column 7, lines 35-48] the group interest profile can be compiled by an average or weighted averages of the interest profiles of each of the contacts, wherein this reinforces the group model. Dudda [column 3, lines 53-64] teaches the suggested recipient of the unpublished post of the use is then provided for display, wherein this performs a function related to at least one media item based on the second preference.)
It would have been obvious to one of ordinary skill in the art at the time of the filing date of the invention to incorporate the teachings of Dudda into the systems of SHIBUI & GERACI. One of ordinary skill in the art would be motivated to provide a system of online sharing which allows users to send, create, and share content with other users. (Dudda [column 1, lines 10-12])
Claim 17 is the method claim corresponding to claim 4 above, therefore is rejected for the same reasons noted previously.
Claim(s) 8 is/are rejected under 35 U.S.C. 103 as being unpatentable over SHIBUI in view of McMahan et al. (US Pub. No. 2019/0340534 A1).
As per Claim 8, McMahan teaches:
acquire, based on the training of the first personalized AI model, a weight related to the first personalized AI model, (McMahan teaches [0079] determining, by a client device, a local model based on one or more local data examples, which is the client side, personalized model training step, and [0028] the client computes an update and a weight/update vale derived directly from the just trained local model’s parameters relative to the prior global model.)
and apply the weight to the main AI model to train the first personalized AI model (McMahan teaches [0029] that each client then sends the update back to the server, where the global update is computed by aggregating all the client side updates, wherein the acquired weight/update is applied to the global model. McMahan teaches [0083] providing the global model to each client device, and receiving the global model, wherein the updated main model is then supplied back to the client, where it seeds the next round of local and personalized training.)
It would have been obvious to one of ordinary skill in the art at the time of the filing date of the invention to incorporate the teachings of McMahan into the system of SHIBUI. One of ordinary skill in the art would be motivated to provide machine learning for more efficient federated learning. (McMahan [0001])
Claim(s) 9 is/are rejected under 35 U.S.C. 103 as being unpatentable over SHIBUI in view of Shen et al. (US Pub. No. 20190026609 A1).
As per Claim 9, Shen teaches:
identify, using the main AI model, an aesthetic score indicating the characteristic of each of the specified media items, (Shen [0039] teaches an aesthetic score generated.)
and train the first personalized AI model using the aesthetic score in addition to the score and the feature (Shen [0042] teaches training a personalized model using the aesthetic score together with a feature.)
It would have been obvious to one of ordinary skill in the art at the time of the filing date of the invention to incorporate the teachings of Shen into the system of SHIBUI. One of ordinary skill in the art would be motivated to provide an accurate system for personalizing an aesthetic score to conserve computational resources by adapting a trained mode using a large data set. (Shen [0003])
Claim(s) 10 is/are rejected under 35 U.S.C. 103 as being unpatentable over SHIBUI in view of GERACI & Dillon et al. (US Pub. No. 2014/0229498 A1).
As per Claim 10, GERACI teaches:
based on determining a group is formed with a plurality of external electronic devices through the communication circuit, acquire an AI model from each of the plurality of external electronic devices, (GERACI teaches [0055, 0121] group formed with external devices; and AI model acquired from each group.)
It would have been obvious to one of ordinary skill in the art at the time of the filing date of the invention to incorporate the teachings of GERACI into the system of SHIBUI & Dillon. One of ordinary skill in the art would be motivated to provide a highly reliable trained personal model even when personal data for training the personal model is insufficient. (GERACI [0005])
Dillon teaches:
identify, using the AI model, one or more preferences of multiple media items related to the group, and based on the one or more preferences, transmit at least one media item from the multiple media items related to the group to the plurality of external electronic devices (Dillon teaches [0004] using the group preferences to select from the database, wherein preferences are identified of the group.)
It would have been obvious to one of ordinary skill in the art at the time of the filing date of the invention to incorporate the teachings of Dillon into the system of SHIBUI, GERACI. One of ordinary skill in the art would be motivated to provide systems for recommending items to members of groups based on the preferences of consumers who are members of the groups. (Dillon [0003])
Allowable Subject Matter
Claims 5-6 & 18-19 are objected to as being dependent upon a rejected base claim, but would be allowable if rewritten in independent form including all of the limitations of the base claim and any intervening claims.
THIS ACTION IS MADE FINAL. 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 nonprovisional extension fee (37 CFR 1.17(a)) 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 mailing date of this final action.
Conclusion
Any inquiry concerning this communication or earlier communications from the examiner should be directed to JOSHUA BULLOCK whose telephone number is (571)270-1395. The examiner can normally be reached 8:00 am - 4:00 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, Kavita Stanley can be reached at 571-272-8352. 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.
/JOSHUA BULLOCK/Primary Examiner, Art Unit 2153 September 12, 2026
Prosecution Insights