DETAILED ACTION
In response to communications filed 24 November 2024, this is the first Office action on the merits. Claims 1-20 are pending.
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 .
Allowable Subject Matter
Claims 4-6, 10, 12, and 19-20 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.
The following is a statement of reasons for the indication of allowable subject matter. Gillick et al., “Learning Dense Representations for Entity Retrieval” and Lin et al. (US 2022/0027569 A1) are the closest prior art on record to the features of claims 4, 10, 12, and 19. However, the features of claims 4, 10, 12, and 19, when considered as a whole in combination with claims 1, 8, 9, 11, and 16 are not taught by Gillick and/or Lin individually or in combination. Accordingly, these features amount to allowable subject matter. Claims 5-6 and 20 likewise contain allowable subject matter based on their dependency.
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.
Claim 8-9, 11, and 13-14 rejected under 35 U.S.C. 101 because the claimed invention is directed to an abstract idea without significantly more. Claim 8 recites
determining target content data;
performing entity word recognition on the target content data to obtain a target entity mention;
encoding the target content data to obtain a target context feature representation corresponding to the target entity mention;
determining, based on a pre-established mapping relationship between an entity mention and an entity in a target knowledge graph, at least one candidate entity corresponding to the target entity mention; obtaining, for each candidate entity, a target semantic feature representation corresponding to the respective candidate entity, the target semantic feature representation being obtained based on an initial semantic feature representation, and the initial semantic feature representation being obtained by encoding candidate semantic feature data of the respective candidate entity using a trained first entity encoding model;
determining a candidate confidence for each candidate entity based on target similarity information between the target context feature representation and the target semantic feature representation of each candidate entity; and
determining a target entity corresponding to the target entity mention from the at least one candidate entity based on the candidate confidence of each candidate entity.
These limitations, under their broadest reasonable interpretation, fall within the Mental Processes grouping of abstract ideas. The steps of “determining,” “performing entity word recognition,” and “encoding” can be performed as mental observations and/or evaluations, perhaps using pen and paper. Accordingly, the claims recite an abstract idea.
This judicial exception is not integrated into a practical application, because the additional limitations of claims 9, 11, and 13-14 likewise fall within the Mental Processes grouping of abstract ideas. Considering the limitations as an ordered combination adds nothing that is not already present when considering the elements individually. There is no indication that the combination of elements improves the functioning of a computer or improves any other technology.
The claims do not include additional elements that are sufficient to amount to significantly more than the judicial exception. Therefore claims 8-9, 11, and 13-14 are not patent eligible.
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 1-2, 7, and 16-17 are rejected under 35 U.S.C. 102(a)(1) as being anticipated by Gillick et al., “Learning Dense Representations for Entity Retrieval.”
Regarding claim 1, Gillick teaches a data processing method, comprising:
receiving a first training sample (see Gillick section 4.2, paragraph 3, “training pair”),
the first training sample including first semantic feature data associated with a first training entity and first training content data associated with the first training entity, the first semantic feature data including first semantic information of the first training entity, and the first training content data including first context information of the first training entity (see Gillick section 4.2, paragraph 3, and 4.1, paragraph 1, “mention-entity pair,” “mentions . . . including their contexts” and “entities . . . including KB [knowledge base] features”);
encoding the first training content data by using a first context encoding model to obtain a first context feature representation corresponding to the first training entity (see Gillick section 4.1, paragraph 1, “mention encoder”);
encoding the first semantic feature data by using a to-be-trained first entity encoding model to obtain a first semantic feature representation corresponding to the first training entity (see Gillick section 4.1, paragraph 1, “entity encoder”);
determining a first feature representation loss based on first similarity information between the first context feature representation and the first semantic feature representation (see Gillick section 4.2, paragraph 3, “For each training pair . . . loss is computed”); and
adjusting first model parameters of the to-be-trained first entity encoding model based on the first feature representation loss to obtain a trained first entity encoding model (see Gillick section 4.1, paragraph 1, “model is trained”).
Regarding claim 16, Gillick teaches an apparatus, comprising:
processing circuitry (see Gillick Table 1, the results shown implicitly teach processing circuitry)
configured to:
receive a first training sample, the first training sample including first semantic feature data associated with a first training entity and first training content data associated with the first training entity, the first semantic feature data including first semantic information of the first training entity, and the first training content data including first context information of the first training entity (see Gillick section 4.2, paragraph 3, and 4.1, paragraph 1, “mention-entity pair,” “mentions . . . including their contexts” and “entities . . . including KB [knowledge base] features”);
encode the first training content data by using a first context encoding model to obtain a first context feature representation corresponding to the first training entity (see Gillick section 4.1, paragraph 1, “mention encoder”);
encode the first semantic feature data by using a to-be-trained first entity encoding model to obtain a first semantic feature representation corresponding to the first training entity (see Gillick section 4.1, paragraph 1, “entity encoder”);
determine a first feature representation loss based on first similarity information between the first context feature representation and the first semantic feature representation (see Gillick section 4.2, paragraph 3, “For each training pair . . . loss is computed”); and
adjust first model parameters of the to-be-trained first entity encoding model based on the first feature representation loss to obtain a trained first entity encoding model (see Gillick section 4.1, paragraph 1, “model is trained”).
Regarding claims 2 and 17, Gillick teaches wherein the first training content data includes first training text containing a first entity mention corresponding to the first training entity; and the encoding the first training content data comprises:
adding boundary markers to the first entity mention in the first training text to obtain first target training text (see Gillick section 4.1, paragraph 3, “mention span is replaced by a special symbol); and
encoding the first target training text using the first context encoding model to obtain the first context feature representation (see Gillick section 4.1, paragraph 1, “mention encoder”).
Regarding claim 7, Gillick teaches further comprising:
obtaining a third training sample including third training content data corresponding to a third training entity, the third training content data corresponding to the third training entity including third context information of the third training entity (see Gillick section 4.2, paragraph 3, and 4.1, paragraph 1, an additional “mention-entity pair” includes third training context data corresponding to a third training entity);
encoding the third training content data corresponding to the third training entity using a to-be-trained third context encoding model to obtain a third context feature representation corresponding to the third training entity (see Gillick section 4.1, paragraph 1, “mention encoder”);
determining a third feature representation loss based on third similarity information between the third context feature representation and a third semantic feature representation corresponding to the third training entity, the third semantic feature representation being obtained by encoding third semantic feature data corresponding to the third training entity by using the trained first entity encoding model (see Gillick section 4.2, paragraph 3, “For each training pair . . . loss is computed”); and
adjusting third model parameters of the to-be-trained third context encoding model based on the third feature representation loss to obtain a trained third context encoding model (see Gillick section 4.1, paragraph 1, “model is trained”).
Claim 8 is rejected under 35 U.S.C. 102(a)(1) and/or 102(a)(2)as being anticipated by Lin et al. (US 2022/0027569 A1).
Regarding claim 8, Lin teaches an entity linking method, comprising:
determining target content data (see Lin [0029], “query information”);
performing entity word recognition on the target content data to obtain a target entity mention (see Lin [0030], a “sequence labeling result,” i.e., target entity mention, is obtained comprising “movie star A1,” “B and C are designated as actors,” and the movie “N,” where [0061] teaches using a “NPL word segmentation tool”);
encoding the target content data to obtain a target context feature representation corresponding to the target entity mention (see Lin [0061], “semantic retrieval word . . . processed by an embedding layer);
determining, based on a pre-established mapping relationship between an entity mention and an entity in a target knowledge graph, at least one candidate entity corresponding to the target entity mention (see Lin [0031], “construct the set of the candidate entity matching the sequence labeling result based on the knowledge graph”);
obtaining, for each candidate entity, a target semantic feature representation corresponding to the respective candidate entity, the target semantic feature representation being obtained based on an initial semantic feature representation, and the initial semantic feature representation being obtained by encoding candidate semantic feature data of the respective candidate entity using a trained first entity encoding model (see Lin [0061], “entity knowledge of the candidate entity . . . processed by an embedding layer,” where the “pre-trained matching model” is a trained first entity encoding model),
determining a candidate confidence for each candidate entity based on target similarity information between the target context feature representation and the target semantic feature representation of each candidate entity (see Lin [0033]-[0034], “performing sematic matching between an entity in the set of the candidate entity and the semantic retrieval part in the sequence labeling result,” where the “semantic relevance” is a candidate confidence); and
determining a target entity corresponding to the target entity mention from the at least one candidate entity based on the candidate confidence of each candidate entity (see Lin [0034], “entity with a high semantic relevance”).
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 3 and 18 are rejected under 35 U.S.C. 103 as being unpatentable over Gillick et al., “Learning Dense Representations for Entity Retrieval” as applied to claim 1 above, and further in view of Zhu et al. (US 11,798,529 B2).
Regarding claims 3 and 18, Gillick teaches wherein the first training content data includes first training data in multiple modals, the first training data in the multiple modals including data selected from first training text (see Gillick section 4.1, paragraph 4, “four text features”); and
the encoding the first training content data comprises:
separately encoding the first training data in the multiple modals using the first context encoding model to obtain first content feature representations respectively corresponding to the multiple modals (see Gillick section 4.1, paragraphs 4 and 6, “mention encoder uses four text features”); and
fusing the first content feature representations respectively corresponding to the multiple modals to obtain the first context feature representation (see Gillick section 4.1, paragraph 6, “embeddings are concatenated”).
Gillick does not explicitly teach at least two types selected from first training text, a first training video, and a first training audio.
However, Zhu teaches at least two types selected from first training text, a first training video, and a first training audio (see Zhu 15:57-61, “acoustic output embeddings”).
It would have been obvious to one of ordinary skill in the art before the effective filing date of the claimed invention to combine training data as taught by Zhu, with the techniques taught by Gillick, “for performing semantic analysis on audio data, audio-visual data, textual data, video data, image data, or other electronic data comprising natural language utterances” (see Zhu 4:6-24).
Claims 11-12 and 15 are rejected under 35 U.S.C. 103 as being unpatentable over Lin et al. (US 2022/0027569 A1) as applied to claim 8 above, and further in view of Gillick et al., “Learning Dense Representations for Entity Retrieval.”.
Regarding claim 11, Lin does not explicitly teach wherein the encoding the target content data comprises: inputting the target content data into a trained third context encoding model; and encoding the target content data using the trained third context encoding model to obtain the target context feature representation corresponding to the target entity mention.
However, Gillick teaches wherein the encoding the target content data comprises: inputting the target content data into a trained third context encoding model; and encoding the target content data using the trained third context encoding model to obtain the target context feature representation corresponding to the target entity mention (see Gillick section 4.1, paragraph 1, “mention encoder”).
It would have been obvious to one of ordinary skill in the art before the effective filing date of the claimed invention to encode the target content data, as taught by Gillick, in combination with the techniques taught by Lin, because “we demonstrate the first accurate, robust, and highly efficient system that is actually a viable substitute for standard, more cumbersome two-stage retrieval and re-ranking systems” (see Gillick section 1, last paragraph).
Regarding claim 12, Lin as modified teaches wherein the trained third context encoding model is obtained by:
receiving a third training sample including third training content data corresponding to a third training entity, the third training content data including third context information of the third training entity (see Gillick section 4.2, paragraph 3, and 4.1, paragraph 1, an additional “mention-entity pair” includes third training context data corresponding to a third training entity);
encoding the third training content data using a to-be-trained third context encoding model to obtain a third context feature representation corresponding to the third training entity (see Gillick section 4.1, paragraph 1, “mention encoder”);
determining a third feature representation loss based on third similarity information between the third context feature representation and a third semantic feature representation corresponding to the third training entity, the third semantic feature representation being obtained by encoding third semantic feature data corresponding to the third training entity using the trained first entity encoding model (see Gillick section 4.2, paragraph 3, “For each training pair . . . loss is computed”); and
adjusting third model parameters of the to-be-trained third context encoding model based on the third feature representation loss to obtain the trained third context encoding model (see Gillick section 4.1, paragraph 1, “model is trained”).
Regarding claim 15, Lin does not explicitly teach wherein the trained first entity encoding model is obtained by:
receiving a first training sample, the first training sample including first semantic feature data associated with a first training entity and first training content data associated with the first training entity, the first semantic feature data including first semantic information of the first training entity, and the first training content data including first context information of the first training entity;
encoding the first training content data by using a first context encoding model to obtain a first context feature representation corresponding to the first training entity;
encoding the first semantic feature data by using a to-be-trained first entity encoding model to obtain a first semantic feature representation corresponding to the first training entity;
determining a first feature representation loss based on first similarity information between the first context feature representation and the first semantic feature representation; and
adjusting first model parameters of the to-be-trained first entity encoding model based on the first feature representation loss.
However, Gillick teaches wherein the trained first entity encoding model is obtained by:
receiving a first training sample, the first training sample including first semantic feature data associated with a first training entity and first training content data associated with the first training entity, the first semantic feature data including first semantic information of the first training entity, and the first training content data including first context information of the first training entity (see Gillick section 4.2, paragraph 3, and 4.1, paragraph 1, “mention-entity pair,” “mentions . . . including their contexts” and “entities . . . including KB [knowledge base] features”);
encoding the first training content data by using a first context encoding model to obtain a first context feature representation corresponding to the first training entity (see Gillick section 4.1, paragraph 1, “mention encoder”);
encoding the first semantic feature data by using a to-be-trained first entity encoding model to obtain a first semantic feature representation corresponding to the first training entity (see Gillick section 4.1, paragraph 1, “entity encoder”);
determining a first feature representation loss based on first similarity information between the first context feature representation and the first semantic feature representation (see Gillick section 4.2, paragraph 3, “For each training pair . . . loss is computed”); and
adjusting first model parameters of the to-be-trained first entity encoding model based on the first feature representation loss (see Gillick section 4.1, paragraph 1, “model is trained”).
It would have been obvious to one of ordinary skill in the art before the effective filing date of the claimed invention to train the first entity encoding model, as taught by Gillick, in combination with the techniques taught by Lin, because “we demonstrate the first accurate, robust, and highly efficient system that is actually a viable substitute for standard, more cumbersome two-stage retrieval and re-ranking systems” (see Gillick section 1, last paragraph).
Conclusion
Any inquiry concerning this communication or earlier communications from the examiner should be directed to Kristopher Andersen whose telephone number is (571)270-5743. The examiner can normally be reached 8:30 AM-5:00 PM ET, Monday-Friday.
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, Ann Lo can be reached at (571) 272-9767. 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.
/Kristopher Andersen/Primary Examiner, Art Unit 2159