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 .
Claim Rejections - 35 USC § 101
35 U.S.C. 101 reads as follows:
Whoever invents or discovers any new and useful process, machine, manufacture, or composition of matter, or any new and useful improvement thereof, may obtain a patent therefor, subject to the conditions and requirements of this title.
Claims 15-20 are rejected under 35 U.S.C. 101 because the claimed invention is directed to non-statutory subject matter.
The claim(s) does/do not fall within at least one of the four categories of patent eligible subject matter because the claims are directed to a computer program product stored on a “computer readable medium,” which encompasses signals per se [Specification, par. 37, disclosing that computer readable media includes computer-readable storage media and computer-readable transmission media (i.e., signals per se)].
It is suggested that the claims be amended to recite a “computer readable storage medium” instead of a “computer readable medium.”
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.
(a)(2) the claimed invention was described in a patent issued under section 151, or in an application for patent published or deemed published under section 122(b), in which the patent or application, as the case may be, names another inventor and was effectively filed before the effective filing date of the claimed invention.
Claims 1-3, 6-10, 13-16, 19, and 20 are rejected under 35 U.S.C. 102(a)(1) and (a)(2) as being anticipated by Ojo et al. (US Pub. 20230297625).
Referring to claim 1, Ojo discloses An apparatus comprising one or more processors and one or more storage devices storing instructions that are operable, when executed by the one or more processors [fig. 13, computing device 1300, processor 1302, memory 1304, storage 1306; par. 157; GNN recommendation system 102 is implemented by computing device 1300], to cause the one or more processors to:
transform graph data of a graphical data structure [fig. 4; pars. 65-68; historical interactions are represented as a set of graphs 400a-400c] into a graph feature set [fig. 4; pars. 65-68; GNN recommendation system generates a digital graph representation based on the historical interactions, the digital graph representation comprising user nodes, attribute nodes, and vis-config nodes], wherein the graphical data structure graphically represents collaboration relationships between component objects [figs. 2 and 4; pars. 44 and 65-68; the set of graphs 400a-400c represents the historical interactions between users and one or more datasets] related to a multi-component system of an application framework [figs. 1 and 4; pars. 38 and 65-68; the users and the datasets are related to a digital content management system], and wherein the graph feature set comprises one or more features associated with the graph data and the component objects [fig. 4; pars. 65-68; note the user nodes, the attributes nodes, and the vis-config nodes, which are associated with the historical interactions, the users, and the datasets];
apply a deep learning model to the graph feature set to generate a graph embedding structure for a portion of the graph data associated with a user identifier [fig. 2; pars. 51-54; the GNN recommendation system analyzes the digital graph representation using a graph neural network (i.e., deep learning techniques) and generates user embeddings, data attribute embeddings, and vis-config embeddings (hereafter “the embeddings”) corresponding to at least an identified target user], wherein the graph embedding structure represents an encoded version of the collaboration relationships for the user identifier [fig. 2; pars. 51-54; the embeddings (e.g., neural network encodings) reflect the historical interactions (including historical interactions for the identified target user)]; and
provide the graph embedding structure to one or more machine learning models configured to generate one or more predictive inferences related to the graph data [fig. 2; pars. 22 and 51-54; the GNN recommendation system processes the embeddings (e.g., utilizing a non-linear function and/or neural network) to determine compatibility scores for possible recommendations (i.e., highly relevant data attributes and/or data visualizations inferred for the identified target user)].
Referring to claim 2, Ojo discloses The apparatus of claim 1, wherein the graph embedding structure comprises a linearized vector data structure that represents the collaboration relationships for the user identifier in a defined format for the one or more machine learning models [pars. 79-81; the GNN recommendation system iteratively applies layers of the graph neural network to generate the embeddings; for the first iteration, the GNN recommendation system uses feature vectors, random vectors or a singular value decomposition of the digital graph representation to initialize the first layer; the GNN recommendation system may also use a function over an adjacency matrix of the digital graph representation to initialize the first layer].
Referring to claim 3, Ojo discloses The apparatus of claim 1, wherein the deep learning model is a graph neural network (GNN) model, and wherein the one or more storage devices store instructions that are operable, when executed by the one or more processors, to further cause the one or more processors to: apply the GNN model to the graph feature set to generate the graph embedding structure [pars. 79-81; note that the graph neural network is used to generate the embeddings].
Referring to claim 6, Ojo discloses The apparatus of claim 1, wherein the one or more storage devices store instructions that are operable, when executed by the one or more processors, to further cause the one or more processors to: receive feedback data from a client device based on the one or more predictive inferences related to the graph data; and update one or more portions of the graph embedding structure based on the feedback data [pars. 59-64; the GNN recommendation system incorporates implicit and explicit feedback (e.g., selecting of a recommendation by a user) over time from client devices of various users to modify and update a graph, the graph neural network, learned parameters, the embeddings, and the recommendations].
Referring to claim 7, Ojo discloses The apparatus of claim 1, wherein the one or more storage devices store instructions that are operable, when executed by the one or more processors, to further cause the one or more processors to: receive feedback data from a client device based on the one or more predictive inferences related to the graph data; and tune one or more weights of the deep learning model based on the feedback data [pars. 51 and 59-64; the GNN recommendation system incorporates implicit and explicit feedback (e.g., selecting of a recommendation by a user) over time from client devices of various users to modify and update a graph, the graph neural network, learned parameters (e.g., weights), the embeddings, and the recommendations].
Referring to claim 8, see the rejection for claim 1, which incorporates the claimed method.
Referring to claim 9, see the rejection for claim 2.
Referring to claim 10, see the rejection for claim 3.
Referring to claim 13, see the rejection for claim 6.
Referring to claim 14, see the rejection for claim 7.
Referring to claim 15, see at least the rejection for claim 1. Ojo further discloses A computer program product, stored on a computer readable medium, comprising instructions that when executed by one or more computers cause the one or more computers to: perform the claimed steps [fig. 13, computing device 1300, processor 1302, memory 1304, storage 1306; par. 157; GNN recommendation system 102 is implemented by computing device 1300].
Referring to claim 16, see the rejection for claim 3.
Referring to claim 19, see the rejection for claim 6.
Referring to claim 20, see the rejection for claim 7.
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.
Claims 4, 11, and 17 are rejected under 35 U.S.C. 103 as being unpatentable over Ojo in view of Verma et al. (US Pub. 20230289610).
Referring to claim 4, Ojo discloses The apparatus of claim 1, wherein the one or more storage devices store instructions that are operable, when executed by the one or more processors, to further cause the one or more processors to: extract one or more entities from the graph data via natural language processing; augment the graph feature set with data associated with the one or more entities to generate an augmented graph features set; and apply the deep learning model to the augmented graph feature set to generate the graph embedding structure [pars. 71 and 72; the GNN recommendation system may generate the digital graph representation as a heterogeneous graph to include additional features or node representations, then use the heterogeneous graph to generate the embeddings].
Ojo does not appear to explicitly disclose that the one or more entities are extracted from the graph data via natural language processing.
However, Verma discloses that the one or more entities are extracted from the graph data via natural language processing [pars. 80 and 84; a data pre-processing engine uses natural language processing algorithms to extract a plurality of graph features based on historical interaction data].
It would have been obvious to one of ordinary skill in the art before the effective filing date of the claimed invention to modify the recommendation system taught by Ojo so that the additional features include graph features extracted using natural language processing algorithms as taught by Verma, with a reasonable expectation of success. The motivation for doing so would have been to provide the graph neural network with graph features from textual datasets, facilitating the ability of the GNN recommendation system to learn from user-level attribute preferences across different datasets [Ojo, par. 73].
Referring to claim 11, see the rejection for claim 4.
Referring to claim 17, see the rejection for claim 4.
Claims 5, 12, and 18 are rejected under 35 U.S.C. 103 as being unpatentable over Ojo in view of Baek et al. (US Pub. 20210089822).
Referring to claim 5, Ojo discloses The apparatus of claim 1, wherein the one or more storage devices store instructions that are operable, when executed by the one or more processors, to further cause the one or more processors to: transform the graph data into an into an adjacency list structure that represents the graph data as a set of linked lists that respectively represent a particular relationship between nodes of the graph data; and generate the graph feature set based on the adjacency list structure [par. 50; the digital graph representation may be implemented as a (adjacency) matrix].
Ojo does not appear to explicitly disclose that the adjacency list structure represents the graph data as a set of linked lists.
However, Baek discloses that the adjacency list structure represents the graph data as a set of linked lists [pars. 88, 160, and 167; user interactions with content items in a content management system (i.e., graph data) are represented as a graph, which may be implemented as a matrix represented using a triplet array representation or a linked list representation].
It would have been obvious to one of ordinary skill in the art before the effective filing date of the claimed invention to modify the recommendation system taught by Ojo so that the digital graph representation of the graph data (i.e., the adjacency matrix) is implemented using a linked list representation as taught by Baek, with a reasonable expectation of success. The motivation for doing so would have been to store the graph data in a compact or compressed form [Baek, par. 88], thereby reducing the amount of storage required to store the graph data.
Referring to claim 12, see the rejection for claim 5.
Referring to claim 18, see the rejection for claim 5.
Conclusion
The following prior art made of record and not relied upon is considered pertinent to applicant's disclosure:
Crisan et al. (US Pub. 20250028939) discloses providing recommendations using graph neural networks, including: obtaining graph data, extracting graph features, generating embeddings, and inferring link predictions.
Contact Information
Any inquiry concerning this communication or earlier communications from the examiner should be directed to GRACE PARK whose telephone number is (571)270-7727. The examiner can normally be reached M-F 8AM-5PM.
Examiner interviews are available via telephone, in-person, and video conferencing using a USPTO supplied web-based collaboration tool. To schedule an interview, applicant is encouraged to use the USPTO Automated Interview Request (AIR) at http://www.uspto.gov/interviewpractice.
If attempts to reach the examiner by telephone are unsuccessful, the examiner’s supervisor, TAMARA KYLE can be reached at (571)272-4241. 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.
/Grace Park/Primary Examiner, Art Unit 2144
