DETAILED ACTION
This action is in response to the reply filed 09 March 2026.
Claims 1–20 are pending. Claims 1, 13, and 18 are independent.
Claims 1–20 are rejected.
Notice of Pre-AIA or AIA Status
The present application, filed on or after 16 March 2013, is being examined under the first inventor to file provisions of the AIA .
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.
Response to Arguments
Applicant’s arguments filed 09 March 2026 have been fully considered but are not persuasive.
Applicant argues that Smith does not teach a query pattern indicative of elements of a query for an entity type associated with an entity (remarks, p. 10). The examiner respectfully disagrees. Smith teaches a knowledge pattern comprising entities of particular types, e.g., a pattern with a disease (Alzheimer’s disease), two genes (CXCR4 and TLR7), and a drug (Rosiglitazone) (Smith, ¶ 25, fig. 1A). The base [knowledge] pattern describes relationships between the different entities (Smith, ¶¶ 26–28) and can be generalized to a generalized base pattern, e.g., Alzheimer’s disease, a set of genes, a second set of genes, and a drug, which can be used to define a query to search for other combinations of entities, e.g., genes and drugs that are related to Alzheimer’s disease and each other, according to the pattern (Smith, ¶¶ 29–32). Furthermore, the claims do not require that the query patterns be derived from historical queries for an entity.
Claim Rejections—35 U.S.C. § 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.
This application currently names joint inventors. In considering patentability of the claims the examiner presumes that the subject matter of the various claims was commonly owned as of the effective filing date of the claimed invention(s) absent any evidence to the contrary. Applicant is advised of the obligation under 37 C.F.R. § 1.56 to point out the inventor and effective filing dates of each claim that was not commonly owned as of the effective filing date of the later invention in order for the examiner to consider the applicability of 35 U.S.C. § 102(b)(2)(C) for any potential 35 U.S.C. § 102(a)(2) prior art against the later invention.
Claims 1–4, 8, 13, and 15–18 are rejected under 35 U.S.C. § 103 as being unpatentable over Gupta et al. (US 2014/0280307 A1) [hereinafter Gupta] in view of Smith et al. (US 2021/0319328 A1) [hereinafter Smith] and Lerman et al. (US 8,560,468 B1) [hereinafter Lerman].
Regarding independent claim 1, Gupta teaches [a] computing system, comprising: a processor; and One or more processors (Gupta, ¶ 83). memory storing instructions that, when executed by the processor, cause the processor to perform acts comprising: Computer-readable media such as memory, storing instructions for execution by the one or more processors (Gupta, ¶ 83). identifying at least one missing fact for an entity referenced in a computer-implemented knowledge graph; Missing data in a knowledge graph is identified (Gupta, ¶¶ 2–5). responsive to identifying the at least one missing fact, generating a query that references an entity based upon an ontology of a knowledge graph and […]; A question [query] is generated based on an entity in a knowledge graph; the query is generated based on the missing data (Gupta, ¶¶ 3, 4, 53). The question/query may be in a natural language or formal language [query pattern (Gupta, ¶ 26). The system generates queries for updating the knowledge graph based on, e.g., English grammar and the missing information in the knowledge graph (Gupta, ¶¶ 59–61). identifying at least one passage from amongst a plurality of passages stored in a passage repository based upon the query; The query is processed by a query processing engine, which may use Internet search engine results, indexes of previously answered questions, etc. [passage repositories] (Gupta, ¶ 25). identifying potential answers to the query in the at least one passage based upon content of the at least one passage and the query; The query processing engine may return a single or multiple answers (Gupta, ¶ 68). […] generating a fact for the entity based upon the answer to the query and the ontology of the computer-implemented knowledge graph; An answer is chosen based on, e.g., a confidence measure (Gupta, ¶ 70). The knowledge graph is updated based on the information received from the query engine, e.g., to update a missing data element in the graph; i.e., the answer from the query system is used to add a new node [fact] to the knowledge graph and connect it to the relevant entity [based on the ontology] (Gupta, ¶ 72). adding the fact to the knowledge graph, wherein the fact is linked to the entity in the knowledge graph; and The knowledge graph is updated using the answer (Gupta, ¶ 72). The knowledge graph comprises entity nodes linked to entity type nodes [facts] representing properties of the entity (Gupta, ¶¶ 33–35). upon receiving a user query that references the entity from a computing device, returning the fact to the computing device based upon the user query. The knowledge graph may be used to answer user queries (Gupta, ¶ 55).
Gupta teaches generating a query for missing information in a knowledge graph, but does not expressly teach using a query pattern as claimed. However, Smith teaches: a query pattern indicative of elements of a query for an entity type associated with the entity A base pattern based on relationships between entities is used to construct a search query for discovering new knowledge patterns (Smith, ¶¶ 28, 52, 54).
It would have been obvious before the effective filing date of the claimed invention to a person having ordinary skill in the art to combine the teachings of Gupta with those of Smith. One would have been motivated to do so in order to increase the efficiency of finding new knowledge (Smith, ¶¶ 2–4).
Gupta/Smith teaches identifying an answer to the query for use in the knowledge graph, but does not expressly teach suppressing invalid answers. However, Lerman teaches: providing the potential answers as input into at least one computer-implemented suppression model configured to suppress invalid answers in the potential answers to the query; A model learns expected categories or values for facts; the model may be used in a system for providing answers to queries to omit or ignore erroneous [invalid] answers that are not in the expected form (Lerman, col. 3 ll. 1–20). obtaining output of the at least one computer-implemented suppression model indicative of an answer to the query; The model provides [outputs] the expected categories and values for the query answering system to use for returning answers (Lerman, col. 3 ll. 1–20).
It would have been obvious before the effective filing date of the claimed invention to a person having ordinary skill in the art to combine the teachings of Gupta/Smith with those of Lerman. One would have been motivated to do so in order to prevent erroneous facts being included in a fact database [i.e., the knowledge graph] (Lerman, col. 1 ll. 5–20).
Regarding dependent claim 2, the rejection of claim 1 is incorporated and Gupta/Smith/Lerman further teaches: wherein the fact comprises a unique identifier for the entity, a predicate that is based upon the query, and the answer. The knowledge graph includes entities, edges representing relationships [predicates], and entity types [answers], e.g., a “California” entity node having a “has capital city” edge connected to a “Sacramento” entity type node (Gupta, ¶¶ 41–44, 53). The nodes of the knowledge graph may have unique identification references [unique identifiers] (Gupta, ¶¶ 38–41).
Regarding dependent claim 3, the rejection of claim 1 is incorporated and Gupta/Smith/Lerman further teaches: wherein the knowledge graph comprises nodes and edges connecting the nodes, wherein the nodes represent entities or attributes, wherein the edges represent relationships between the entities or relationships between the entities and the attributes. The knowledge graph comprises nodes connected by edges, with the nodes representing entities or entity types [entity attributes] and the edges representing relationships between the nodes (Gupta, ¶¶ 41–44).
Regarding dependent claim 4, the rejection of claim 1 is incorporated and Gupta/Smith/Lerman further teaches: wherein identifying that the at least one missing fact for the entity is not present in the knowledge graph is based upon an ontology for the entity type associated with the entity. The question is generated based on identifying that information is missing from the knowledge graph, e.g., by comparing an entity to a schema table (Gupta, ¶¶ 2–5, 66).
Regarding dependent claim 8, the rejection of claim 1 is incorporated and Gupta/Smith/Lerman further teaches: wherein the invalid answers are suppressed using at least one of: regular expression matching; Attributes and values may be mapped to regular expressions (Lerman, col. 9 ll. 35–40). part of speech analysis; or dependency tree analysis.
Regarding independent claim 13, this claim recites limitations similar to those of claim 1, and therefore is rejected for the same reasons.
Regarding dependent claim 15, the rejection of claim 13 is incorporated and Gupta/Smith/Lerman further teaches: wherein the answer is one of an attribute; or The nodes include entity type nodes, e.g., defining a type or other attribute of the entity (Gupta, ¶¶ 44–49). The answer fills in missing information for an entity, i.e., one of the attributes, such as the architect of a building entity (Gupta, ¶ 59). an identifier for a second entity, wherein the second entity is referenced in the knowledge graph.
Regarding dependent claim 16, the rejection of claim 13 is incorporated and Gupta/Smith/Lerman further teaches: wherein the plurality of passages include web pages that comprise unstructured text. Query processing includes Internet search engine results (Gupta, ¶ 25).
Regarding dependent claim 17, the rejection of claim 13 is incorporated and Gupta/Smith/Lerman further teaches: wherein the fact is based upon a type of the entity. The attributes of a node [fact] may depend on what the node represents, e.g., a “movie” entity may have connected nodes for “actor” and “director” (Gupta, ¶¶ 41–43).
Regarding independent claim 18, this claim recites limitations similar to those of claim 4, and therefore is rejected for the same reasons. Gupta further teaches a computer-readable medium in ¶ 83.
Claim 5 is rejected under 35 U.S.C. § 103 as being unpatentable over Gupta et al. (US 2014/0280307 A1) [hereinafter Gupta] in view of Smith et al. (US 2021/0319328 A1) [hereinafter Smith] and Lerman et al. (US 8,560,468 B1) [hereinafter Lerman], further in view of Raiskin (US 2021/0311728 A1).
Regarding dependent claim 5, the rejection of claim 1 is incorporated. Gupta/Smith/Lerman teaches identifying answers to a question, but does not expressly teach using recall or precision passage ranking models. However, Raiskin teaches: wherein the at least one passage is identified based upon: The position of an element [passage] on a webpage is identified (Raiskin, ¶¶ 12–14). The position may be identified using an ensemble model including two classification models (Raiskin, ¶ 25). a recall passage ranking model that identifies a first subset of the plurality of passages based upon the query; and The first model is a model that focuses on recall [recall passage ranking model] (Raiskin, ¶¶ 25–26). a precision passage ranking model that identifies a second subset of the plurality of passages based upon the query, wherein a number of passages in the first subset is greater than a number of passages in the second subset. The second model is a model that focuses on precision [precision passage ranking model]; the second model refines the results of the first model, i.e., the first model’s result set is larger than the second model’s result set (Raiskin, ¶ 25–26). The output of the model is a likelihood [ranking] that each element is associated with particular content, e.g., a specific payment method (Raiskin, claims 6 and 13).
It would have been obvious before the effective filing date of the claimed invention to a person having ordinary skill in the art to combine the teachings of Gupta/Smith/Lerman with those of Raiskin. One would have been motivated to do so in order to more accurately predict the type of content associated with a position on the page (Raiskin, ¶¶ 25–26).
Claim 6 is rejected under 35 U.S.C. § 103 as being unpatentable over Gupta et al. (US 2014/0280307 A1) [hereinafter Gupta] in view of Smith et al. (US 2021/0319328 A1) [hereinafter Smith] and Lerman et al. (US 8,560,468 B1) [hereinafter Lerman], further in view of Bakshi et al. (US 2017/0068683 A1) [hereinafter Bakshi].
Regarding dependent claim 6, the rejection of claim 1 is incorporated. Gupta/Smith/Lerman teaches generating a question/query but does not expressly teach using a query pattern from query logs. However, Bakshi teaches: wherein the query pattern is mined from query logs of a search engine. A query pattern for a query is selected from a query pattern database (Bakshi, ¶ 4). The query pattern database stores query patterns extracted [mined] from query logs (Bakshi, ¶¶ 20, 50).
It would have been obvious before the effective filing date of the claimed invention to a person having ordinary skill in the art to combine the teachings of Gupta/Smith/Lerman with those of Bakshi. One would have been motivated to do so in order to generate more effective search queries by using more relevant search query patterns (Bakshi, ¶ 66).
Claims 7 and 20 are rejected under 35 U.S.C. § 103 as being unpatentable over Gupta et al. (US 2014/0280307 A1) [hereinafter Gupta] in view of Smith et al. (US 2021/0319328 A1) [hereinafter Smith] and Lerman et al. (US 8,560,468 B1) [hereinafter Lerman], further in view of Huang et al. (US 2020/0117742 A1) [hereinafter Huang].
Regarding dependent claim 7, the rejection of claim 1 is incorporated and Gupta/Smith/Lerman further teaches: searching a web index based upon the user query; and The facts may be from web sources (Lerman, col. 5 ll. 30–40).
Gupta/Smith/Lerman teaches obtaining facts from web sources, but does not expressly teach displaying a search results page. However, Huang teaches: identifying uniform resource locators (URLs) based upon search results for the search, wherein a search engine results page that includes the URLs and the fact is returned to the computing device, wherein the search results page is presented on a display. A search results page is displayed with a query answer or entity description [facts] in curated position(s) together with search results (Huang, ¶¶ 13–14). The search results include main results, such as web search results (Huang, ¶ 15). The search results may be displayed with a title, link URL, and summary or snippet (Huang, ¶ 19).
It would have been obvious before the effective filing date of the claimed invention to a person having ordinary skill in the art to combine the teachings of Gupta/Smith/Lerman with those of Huang. One would have been motivated to do so in order to make it easier for the user to view the source of the fact (Huang, ¶ 1).
Regarding dependent claim 20, the rejection of claim 18 is incorporated. Gupta/Smith/Lerman teaches outputting information from the knowledge graph (Gupta, ¶ 62) but does not expressly teach outputting as audible words using a speaker. However, Huang teaches: wherein a speaker of the computing device emits audible words that are indicative of the fact. The computer may receive queries via natural language speech at a microphone, and output results via speech audio using a speaker (Huang, ¶ 12).
It would have been obvious before the effective filing date of the claimed invention to a person having ordinary skill in the art to combine the teachings of Gupta/Smith/Lerman with those of Huang. Doing so would have been a matter of simple substitution of one known element [the output method of Gupta] with another [the output method of Huang] to produce a predictable results [outputting facts from a knowledge graph as words from a speaker].
Claim 9 is rejected under 35 U.S.C. § 103 as being unpatentable over Gupta et al. (US 2014/0280307 A1) [hereinafter Gupta] in view of Smith et al. (US 2021/0319328 A1) [hereinafter Smith] and Lerman et al. (US 8,560,468 B1) [hereinafter Lerman], further in view of Forsyth et al. (US 2021/0365453 A1) [hereinafter Forsyth].
Regarding dependent claim 9, the rejection of claim 1 is incorporated. Gupta/Smith/Lerman teaches using a question/answer system to add missing data to a knowledge graph, but does not expressly teach normalizing the answer to a format of the knowledge graph. However, Forsyth teaches: subsequent to identifying the potential answers to the query and prior to generating the fact, normalizing the potential answers to a format supported by the knowledge graph, wherein the answer is identified based upon the answer being successfully normalized to the format supported by the knowledge graph. A controller receives response data from one or more data sources, which is formatted/normalized such that the response data may be used in a data structure, e.g., a knowledge graph (Forsyth, ¶ 52). The response data is in response to a query generated by an investigation input (Forsyth, ¶ 47). The controller augments or updates the data structure [knowledge graph] with the response data after the reformatting/normalization (Forsyth, ¶ 53).
It would have been obvious before the effective filing date of the claimed invention to a person having ordinary skill in the art to combine the teachings of Gupta/Smith/Lerman with those of Forsyth. One would have been motivated to do so in order to reduce the amount of manual work required to construct the knowledge graph, and increase the amount of data available for constructing the graph, i.e., by providing automatic reformatting of data from a plurality of different data sources (Forsyth, ¶¶ 5, 52, 53).
Claim 10 is rejected under 35 U.S.C. § 103 as being unpatentable over Gupta et al. (US 2014/0280307 A1) [hereinafter Gupta] in view of Smith et al. (US 2021/0319328 A1) [hereinafter Smith] and Lerman et al. (US 8,560,468 B1) [hereinafter Lerman], further in view of Hassanzadeh et al. (US 2017/0124217 A1) [hereinafter Hassanzadeh].
Regarding dependent claim 10, the rejection of claim 1 is incorporated. Gupta/Smith/Lerman teaches determining an answer from multiple answers, but does not expressly teach comparing the fact/answer to a second fact/answer already in the knowledge graph for consistency. However, Hassanzadeh teaches: subsequent to generating the fact and prior to adding the fact to the knowledge graph, comparing the fact to a second fact for the entity in the knowledge graph, wherein the fact is added to the knowledge graph upon determining that the fact and the second fact are consistent. A knowledge graph is augmented with an attribute, based on a consistency ranking that ranks attributes based in part on how often the attribute occurs with an attribute that is already in the knowledge graph (Hassanzadeh, ¶ 56).
It would have been obvious before the effective filing date of the claimed invention to a person having ordinary skill in the art to combine the teachings of Gupta/Smith/Lerman with those of Hassanzadeh. One would have been motivated to do so in order to ensure that the best attributes are used for augmenting the knowledge graph (Hassanzadeh, ¶ 56).
Claims 11, 12, and 14 are rejected under 35 U.S.C. § 103 as being unpatentable over Gupta et al. (US 2014/0280307 A1) [hereinafter Gupta] in view of Smith et al. (US 2021/0319328 A1) [hereinafter Smith] and Lerman et al. (US 8,560,468 B1) [hereinafter Lerman], further in view of Brennan et al. (US 2018/0075359 A1) [hereinafter Brennan].
Regarding dependent claim 11, the rejection of claim 1 is incorporated. Gupta/Smith/Lerman teaches determining an answer from multiple answers, but does not expressly teach using a deep learning model. However, Brennan teaches: subsequent to generating the fact and prior to adding the fact to the knowledge graph, providing the fact and the at least one passage as input to a deep learning model, wherein the fact is added to the knowledge graph upon the deep learning model determining that the fact is consistent with the at least one passage. A knowledge graph is expanded by generating questions and presenting the questions to a human analyst or a cognitive system (Brennan, ¶ 21). The correctness of the answer may be determined using an analysis of the answer and an evidence passage (Brennan, ¶ 115). The analysis a deep analysis using reasoning algorithms (Brennan, ¶ 113). The analysis and other functions use IBM Watson/DeepQA (Brennan, ¶ 3).
It would have been obvious before the effective filing date of the claimed invention to a person having ordinary skill in the art to combine the teachings of Gupta/Smith/Lerman with those of Brennan. One would have been motivated to do so in order to increase the likelihood of selecting the correct answer/fact to add to the knowledge graph (Brennan, ¶ 113).
Regarding dependent claim 12, the rejection of claim 1 is incorporated. Gupta/Smith/Lerman teaches generating answers to a question, but does not expressly teach determining that an entity in a passage matches that of the question/query based on the knowledge graph. However, Brennan teaches: wherein the at least one passage references the entity, the acts further comprising: The evidence passage for a candidate answer may have a term [entity] that matches that of the question (Brennan, ¶¶ 114–115). determining that the entity referenced in the at least one passage matches the entity referenced in the query based upon an entry in the knowledge graph for the entity. A knowledge graph is expanded by generating questions and presenting the questions to a human analyst or a cognitive system (Brennan, ¶ 21). The cognitive system may use data already in the knowledge graph, as well as the question, when generating an answer for expanding the knowledge graph (Brennan, ¶¶ 77–79).
It would have been obvious before the effective filing date of the claimed invention to a person having ordinary skill in the art to combine the teachings of Gupta/Smith/Lerman with those of Brennan. One would have been motivated to do so in order to increase the likelihood of selecting the correct hypotheses [answer/fact] to add to the knowledge graph (Brennan, ¶¶ 79, 80).
Regarding dependent claim 14, the rejection of claim 13 is incorporated. Gupta/Smith/Lerman teaches identifying answers to questions, but does not expressly teach a machine reading comprehension model. However, Brennan teaches: wherein the potential answers are identified based upon a machine reading comprehension model that takes the at least one passage and the query as input and that outputs the potential answers based upon the input. Answers to an input question are generated using a QA pipeline [machine reading comprehension model] performing a deep analysis of the language of the question and a corpus of data, including reasoning algorithms, such as natural language analysis, lexical analysis, etc. in order to generate candidate answers (Brennan, ¶ 59).
It would have been obvious before the effective filing date of the claimed invention to a person having ordinary skill in the art to combine the teachings of Gupta/Smith/Lerman with those of Brennan. One would have been motivated to do so in order to generate more relevant answers based on a larger amount and/or variety of data (Brennan, ¶¶ 2–3).
Claim 19 is rejected under 35 U.S.C. § 103 as being unpatentable over Gupta et al. (US 2014/0280307 A1) [hereinafter Gupta] in view of Smith et al. (US 2021/0319328 A1) [hereinafter Smith] and Lerman et al. (US 8,560,468 B1) [hereinafter Lerman], further in view of Srinivasan et al. (US 2019/0294732 A1) [hereinafter Srinivasan].
Regarding dependent claim 19, the rejection of claim 18 is incorporated. Gupta/Smith/Lerman teaches adding missing data to a knowledge graph, but does not expressly teach a domain-specific graph for an organization. However, Srinivasan teaches: wherein the knowledge graph is a domain-specific knowledge graph for an organization. A knowledge graph is generated for a specific domain, which may be an enterprise [organization] (Srinivasan, ¶¶ 1–6).
It would have been obvious before the effective filing date of the claimed invention to a person having ordinary skill in the art to combine the teachings of Gupta/Smith/Lerman with those of Srinivasan. One would have been motivated to do so in order to make the knowledge graph more useful for enterprise use, i.e., by including enterprise-specific or domain-specific data not found in knowledge graphs generated using generic corpora (Srinivasan, ¶ 2).
Conclusion
THIS ACTION IS MADE FINAL. Applicant is reminded of the extension of time policy as set forth in 37 C.F.R. § 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 C.F.R. § 1.17(a)) pursuant to 37 C.F.R. § 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.
The prior art made of record and not relied upon is considered pertinent to Applicant's disclosure.
Any inquiry concerning this communication or earlier communications from the examiner should be directed to Tyler Schallhorn whose telephone number is 571-270-3178. The examiner can normally be reached Monday through Friday, 8:30 a.m. to 6 p.m. (ET).
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 on 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 the USA or Canada) or 571-272-1000.
/Tyler Schallhorn/Examiner, Art Unit 2144
/TAMARA T KYLE/Supervisory Patent Examiner, Art Unit 2144
Prosecution Insights