DETAILED ACTION
This Office action is in response to Applicant’s reply filed 05/13/2026.
Claims 1-2 and 4-21 are pending.
Claims 1, 4; 8; and 15 are amended.
Claim 3 is canceled. Claim 21 is new.
Claims 1-2 and 4-21 are rejected.
Notice of 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 .
Information Disclosure Statement
The information disclosure statement (IDS) submitted on 05/13/2026 was filed after the mailing date of the non-final Office Action on 02/13/2026. The submission is in compliance with the provisions of 37 CFR 1.97. Accordingly, the information disclosure statement is being considered by the examiner.
Response to Arguments
35 U.S.C. 101
Applicant’s arguments, see p9, filed 05/13/2026, with respect to claims 1-2 and 4-21 have been fully considered and are persuasive. The 35 U.S.C. 101 rejection of claims 1-2 and 4-21 has been withdrawn.
35 U.S.C. 103
Applicant’s arguments, see pp10-11, filed 05/13/2026, with respect to the rejection(s) of claim(s) 1-2, 4-6, 8-9, 11-13, 15-16, and 18-19 under 35 U.S.C. 103 have been fully considered and are persuasive. Therefore, the rejection has been withdrawn. However, upon further consideration, a new ground(s) of rejection is made under 35 U.S.C. 103 as being unpatentable over Tripathi in view of Stubley in further view of newly incorporated reference Roberts.
Statutory Review under 35 USC § 101
Claims 1-7 are directed towards a method and have been reviewed.
Claims 1-7 appear to be directed to significantly more than an abstract idea based on the subject matter eligibility consideration.
Claims 8-14 are directed toward an article of manufacture and have been reviewed.
Claims 8-14 initially appear to be statutory, as the article of manufacture excludes transitory signals (claim says non-transitory).
Claims 8-14 also appear to be directed to significantly more than an abstract idea based on the subject matter eligibility consideration.
Claims 15-20 are directed toward a system and have been reviewed.
Claims 15-20 initially appear to be statutory as the system contains hardware.
Claims 15-20 also appear to be directed to significantly more than an abstract idea based on the subject matter eligibility consideration.
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 1-2, 5-6; 8-9, 11-13; 15-16, and 18-19 are rejected under 35 U.S.C. 103 as being unpatentable over Tripathi et al., U.S. Patent Application Publication No. 2019/0005124 (hereinafter Tripathi) in view of Stubley et al., U.S. Patent Application Publication No. 2016/0179934 (hereinafter Stubley) in further view of Roberts et al., U.S. Patent Application Publication No. 2015/0324454 (hereinafter Roberts).
Regarding claim 1, Tripathi teaches:
A method comprising: receiving, via a graphical user interface (GUI) presented via a user device, an input query that is associated with a research topic and that includes a first concept … and a second concept … the research topic being associated with an inquiry regarding whether a relationship exists between the first concept and the second concept; (Tripathi FIG. 1, FIG. 5, ¶ 0124: At a step 502, a first input area, for receiving the search query from the user, is provided on a first user-interface of the computing device. At a step 504, one or more query segments and relations between the one or more query segments are displayed on a second area of the first user-interface. The received search query is developed to obtain the one or more query segments and the relations between the one or more query segments ... a list of one or more concepts [relevant to research topic] associated with the extracted search results)
performing, based at least in part on the text of the input query in the machine language, a search of a plurality of knowledge data sources that each include a plurality of documents; (Tripathi FIG. 1, FIG. 5, ¶ 0124: At a step 506, an arranged set of extracted search results is provided in a third area of the first user-interface. The search results are extracted, from at least one database [shows a plurality of knowledge data sources], based on the developed search query and arranged based on one or more parameters associated with the extracted search results ... the information related to the selected search result comprises a summary of the selected search result, one or more documents associated with the selected search result [shows plurality of documents])
…
generating a query result … that represents the connection between the first concept and the second concept; and (Tripathi FIG. 1, FIG. 5, ¶ 0124: The received search query is developed to obtain the one or more query segments and the relations between the one or more query segments [shows connection between concepts]. At a step 506, an arranged set of extracted search results is provided in a third area of the first user-interface. The search results are extracted, from at least one database, based on the developed search query [shows generating])
visually displaying, via the GUI presented via the user device, the query result. (Tripathi FIG. 1, FIG. 5, ¶ 0124: At step 510, information related to a selected search result or a selected concept, is provided on a second user-interface, in response to a selection input, from the user, based on one of a search result or a concept associated with the search results. Moreover, the information related to the selected search result comprises a summary of the selected search result, one or more documents associated with the selected search result, and concepts associated with the selected search result. Furthermore, the information related to the selected concept comprises a summary of the selected concept, and one or more documents associated with the selected concept)
Tripathi does not expressly disclose a first concept that serves as a starting point for the research topic and a second concept that serves as an intermediary point or an ending point of the research topic.
Tripathi further does not expressly disclose:
parsing, using a semantic parser, text of the input query to convert the text of the input query into a machine language that is a machine-understandable representation of a meaning of the text;
performing, based at least in part on the text of the input query in the machine language, a search of a plurality of knowledge data sources that each include a plurality of documents;
translating, using a natural language understanding (NLU) engine, the query result into a machine-readable structured representation of the query result; and
visually displaying, based at least in part on translating the query result and via the GUI presented via the user device, the query result.
identifying, based at least in part on the search, at least one evidence passage that serves as at least one potential link between the first concept and the second concept;
analyzing, using natural language processing, the at least one evidence passage to determine that information included within the at least one evidence passage represents a connection between the first concept and the second concept;
Tripathi further does not expressly disclose generating a query result indicating the at least one evidence passage.
However, Stubley addresses this by teaching:
identifying, based at least in part on the search, at least one evidence passage that serves as at least one potential link between the first concept and the second concept; (Stubley ¶ 0078-0081: the result of applying one or more search queries constructed by query builder 170 to unstructured data set(s) 140 may be a set of documents and/or passages identified as relevant to a search query and passed to evidence scorer 180. The set of returned documents/passages may be thresholded for relevance in any suitable way ... evaluate the retrieved passages in terms of the strength of supporting evidence that they provide for the extracted answer information as contributing to the best answer to the user's question)
analyzing, using natural language processing, the at least one evidence passage to determine that information included within the at least one evidence passage represents a connection between the first concept and the second concept; (Stubley FIG. 5, ¶ 0124: At act 530, documents containing natural language text may be analyzed, including analyzing one or more passages of text in the documents to determine whether the passage entails any of the hypotheses from the question. Exemplary techniques for entailment analysis are described above. At act 540, in response to determining that a passage entails a question hypothesis, the passage may be identified as providing supporting evidence for the generated answer to the question; Stubley ¶ 0084-0086: the strength of a passage's supporting evidence for answer information for a question portion may be evaluated by determining whether the passage entails a hypothesis corresponding to the question portion ... evidence scorer 180 may score passages based at least in part on the degree to which they entail one or more question hypotheses, and may rank passages based on their scores ... an entailment/contradiction score computed as described above ... may be used as one feature input to a statistical classifier used to score the strength of a passage's supporting evidence for an answer item)
Stubley further teaches generating a query result indicating the at least one evidence passage. (Stubley FIG. 5, ¶ 0123-0124: At act 550, the answer and the passage(s) identified as providing supporting evidence for that answer may be presented to the user in response to the input question. In some embodiments, as discussed above, multiple different passages may be scored based at least in part on the strength of the passages' supporting evidence for the answer to the question, and one or more of the passages may be selected for presentation to the user based on the passages' scores)
It would have been obvious to one of ordinary skill in the art before the effective filing date of the claimed invention to combine the functioning of the evidence passage retrieval of Tripathi with the evidence passage retrieval of Stubley.
In addition, both of the references (Tripathi and Stubley) disclose features that are directed to analogous art, and they are directed to the same field of endeavor, such as question answering interfaces.
Motivation to do so would be to improve the functioning of Tripathi returning search result documents in response to compound queries with the functioning in similar reference Stubley also returning search result documents in response to compound queries but with the improvement of scoring techniques.
Motivation to do so would also be the teaching, suggestion, or motivation for a person of ordinary skill in the art to implement designing a QA system to make intelligent use of both structured and unstructured data sources in its knowledge base as seen in Stubley ¶ 0028.
Tripathi in view of Stubley does not expressly disclose:
receiving, via a graphical user interface (GUI) presented via a user device, an input query that is associated with a research topic and that includes a first concept that serves as a starting point for the research topic and a second concept that serves as an intermediary point or an ending point of the research topic,
parsing, using a semantic parser, text of the input query to convert the text of the input query into a machine language that is a machine-understandable representation of a meaning of the text;
performing, based at least in part on the text of the input query in the machine language, a search of a plurality of knowledge data sources that each include a plurality of documents;
translating, using a natural language understanding (NLU) engine, the query result into a machine-readable structured representation of the query result; and
visually displaying, based at least in part on translating the query result and via the GUI presented via the user device, the query result.
However, Roberts taches:
receiving, via a graphical user interface (GUI) presented via a user device, an input query that is associated with a research topic and that includes a first concept that serves as a starting point for the research topic and a second concept that serves as an intermediary point or an ending point of the research topic, (Roberts FIG. 3, ¶ 0071: there is a hierarchy of information going from topics to entities to subtopics. As used herein, the term topic refers to any subject matter of interest, and the term “entity” refers to a particular type of topic having strongly typed attributes that distinguish it from other entities. As with topics, a subtopic does not imply any particular structure (although it may have structure), but it is expressly predicated on being about an entity; Roberts ¶ 0085-0097, see ¶ 0086: The first document 302 may generally include data relating to a topic identified by a topic identifier 310 ... the passage 312 may include a variety of textural references such as an entity identifier 314, an entity mention 316, and a citation 318; see ¶ 0087: An entity identifier 314 may include any mechanism for referring to a particular entity, e.g., an entity identified by the entity identifier 320 that is the focus of the entity profile 308 included in the second document 304; see ¶ 0093: the second document 304 may include a number of attributes 324, 328 and subtopics 340 used to describe the entity ... the attributes in the second document 304 may include one or more properties, traits, characteristics, details, features, relationships, and so forth, of an entity, and various schemas or folksonomies may capture such attributes 324 in a representation of the entity)
parsing, using a semantic parser, text of the input query to convert the text of the input query into a machine language that is a machine-understandable representation of a meaning of the text; (Roberts ¶ 0036: the server 106 may support search activity by periodically searching for content at remote locations on the data network 102 and indexing any resulting content for subsequent search by a client 104. This may include storing location or address information for a particular document as well as parsing the document in any suitable manner to identify words, images, media, metadata and the like, as well as the creation of feature vectors or other derivative data [relevant to machine-understandable representation] to assist in similarity-type comparisons, dissimilarity comparisons, or other analysis; show similarity-type comparisons used for querying in Roberts FIG. 6, ¶ 0159-0160: The extension may also show recommendations 614 from a recommendation engine that analyzes the documents bookmarked by the user and suggests other documents that are similar and may be helpful to the user ... the recommendation engine presents the recommendation 616 to the user; see machine-understandable representation of a meaning of the text through ¶ 0062-0066: techniques may be used to derive metadata that characterizes the content in more abstract forms. Thus for example the fifth data representation 226 may be a feature vector or feature collection constructed automatically from the semistructured output 206 of a tagger process 212. The purpose of a feature vector is to automatically capture aspects of a piece of unstructured or semistructured data in a form that is easily compared to other pieces of data... Feature vectors for text typically capture words and phrases and concepts derived from the words and phrases in a document)
performing, based at least in part on the text of the input query in the machine language, a search of a plurality of knowledge data sources that each include a plurality of documents; (Roberts ¶ 0036: the server 106 may support search activity by periodically searching for content at remote locations on the data network 102 [relevant to plurality of knowledge data sources] and indexing any resulting content for subsequent search by a client 104. This may include storing location or address information for a particular document as well as parsing the document in any suitable manner to identify words, images, media, metadata and the like, as well as the creation of feature vectors or other derivative data to assist in similarity-type comparisons, dissimilarity comparisons, or other analysis; Roberts FIG. 4, ¶ 0122-0123: The search results 418 may be located and retrieved for the target entity based on the entity profile 408 in the first window 402 using an algorithm 420 to traverse a data network or other source(s) of data ... The algorithm 420 may use the entity profile 408 as a query for retrieving, ranking, and displaying the search results 418. One possible approach to implementing the algorithm 420 is to construct a feature vector for the profile 408; see also Roberts ¶ 0005: the second window shows a plurality of search results from one or more sources that identify documents each containing a mention with a predetermined likelihood of referring to the entity)
translating, using a natural language understanding (NLU) engine, the query result into a machine-readable structured representation of the query result; and (Roberts FIG. 4, ¶ 0122: The second window 404 may be concurrently visible with the first window 402, and may display a plurality of search results 418 [shows a structured representation] … The search results 418 may be displayed with a preview of pertinent information included in the particular search result 418, such as highlighting or otherwise emphasizing keywords or other relevant text within a window of surrounding text [relevant to natural language understanding], where the keywords are specified by the user or determined by the algorithm 420; see more natural language understanding in at least Roberts ¶ 0088: The context surrounding an entity mention 316 may include a larger span of natural language prose near the entity mention 316 substring; see entity mentions in the query result in Roberts ¶ 0128: The third window 406 may include highlighting or other visual emphasis of mentions of the target entity 410 in the selected search result 428)
visually displaying, based at least in part on translating the query result and via the GUI presented via the user device, the query result. (Roberts FIG. 6, ¶ 0159-0160: The extension may also show recommendations 614 from a recommendation engine that analyzes the documents bookmarked by the user and suggests other documents that are similar and may be helpful to the user ... One of the recommendations 616 may be generated from analyzing a batch of documents related to the selected item in the foldering tree and finding common properties of those documents, such as long strings in common across the texts. After finding such a common property, the recommendation engine presents the recommendation 616 to the user; see also Roberts FIG. 4, ¶ 0116-0123, notably ¶ 0122: The second window 404 may be concurrently visible with the first window 402, and may display a plurality of search results 418 ... The search results 418 may be displayed with a preview of pertinent information included in the particular search result 418, such as highlighting or otherwise emphasizing keywords or other relevant text within a window of surrounding text, where the keywords are specified by the user or determined by the algorithm 420)
It would have been obvious to one of ordinary skill in the art before the effective filing date of the claimed invention to combine the functioning of the evidence passage retrieval of Tripathi as modified with the passage retrieval of Roberts.
In addition, both of the references (Tripathi as modified and Roberts) disclose features that are directed to analogous art, and they are directed to the same field of endeavor, such as textual data retrieval and display.
Motivation to do so would be to improve the functioning of Tripathi as modified returning search result documents in response to queries with the functioning in similar reference Roberts also returning search result documents in response to queries but with the improvement of in-progress entity profiles.
Motivation to do so would also be the teaching, suggestion, or motivation for a person of ordinary skill in the art to more quickly build a comprehensive description of an entity of interest as seen in Roberts ¶ 0004.
Regarding claim 8, Tripathi teaches:
One or more non-transitory computer-readable media storing one or more computer-executable instructions that, when executed by one or more processors, cause the one or more processors to perform operations comprising: (Tripathi ¶ 0002: the present disclosure also relates to computer readable medium containing program instructions for execution on a computer system, which when executed by a computer, cause the computer to perform method steps for presenting information related to search; Tripathi ¶ 0092: Examples of computing device include, but are not limited to, cellular phones, personal digital assistants (PDAs), handheld devices, wireless modems, laptop computers, personal computers, etc. Additionally, the computing device includes a casing, a memory, a processor)
receiving, via a graphical user interface (GUI) presented via a user device, an input query that is associated with a research topic and that includes a first concept … and a second concept … the research topic being associated with an inquiry regarding whether a relationship exists between the first concept and the second concept; (Tripathi FIG. 1, FIG. 5, ¶ 0124: At a step 502, a first input area, for receiving the search query from the user, is provided on a first user-interface of the computing device. At a step 504, one or more query segments and relations between the one or more query segments are displayed on a second area of the first user-interface. The received search query is developed to obtain the one or more query segments and the relations between the one or more query segments ... a list of one or more concepts [relevant to research topic] associated with the extracted search results)
performing, based at least in part on the text of the input query in the machine language, a search of a plurality of knowledge data sources that each include a plurality of documents; (Tripathi FIG. 1, FIG. 5, ¶ 0124: At a step 506, an arranged set of extracted search results is provided in a third area of the first user-interface. The search results are extracted, from at least one database [shows a plurality of knowledge data sources], based on the developed search query and arranged based on one or more parameters associated with the extracted search results ... the information related to the selected search result comprises a summary of the selected search result, one or more documents associated with the selected search result [shows plurality of documents])
…
generating a query result … that represents the connection between the first concept and the second concept; and (Tripathi FIG. 1, FIG. 5, ¶ 0124: The received search query is developed to obtain the one or more query segments and the relations between the one or more query segments [shows connection between concepts]. At a step 506, an arranged set of extracted search results is provided in a third area of the first user-interface. The search results are extracted, from at least one database, based on the developed search query [shows generating])
visually displaying, via the GUI presented via the user device, the query result. (Tripathi FIG. 1, FIG. 5, ¶ 0124: At step 510, information related to a selected search result or a selected concept, is provided on a second user-interface, in response to a selection input, from the user, based on one of a search result or a concept associated with the search results. Moreover, the information related to the selected search result comprises a summary of the selected search result, one or more documents associated with the selected search result, and concepts associated with the selected search result. Furthermore, the information related to the selected concept comprises a summary of the selected concept, and one or more documents associated with the selected concept)
Tripathi does not expressly disclose a first concept that serves as a starting point for the research topic and a second concept that serves as an intermediary point or an ending point of the research topic.
Tripathi further does not expressly disclose:
parsing, using a semantic parser, text of the input query to convert the text of the input query into a machine language that is a machine-understandable representation of a meaning of the text;
performing, based at least in part on the text of the input query in the machine language, a search of a plurality of knowledge data sources that each include a plurality of documents;
translating, using a natural language understanding (NLU) engine, the query result into a machine-readable structured representation of the query result; and
visually displaying, based at least in part on translating the query result and via the GUI presented via the user device, the query result.
identifying, based at least in part on the search, at least one evidence passage that serves as at least one potential link between the first concept and the second concept;
analyzing, using natural language processing, the at least one evidence passage to determine that information included within the at least one evidence passage represents a connection between the first concept and the second concept;
Tripathi further does not expressly disclose generating a query result indicating the at least one evidence passage.
However, Stubley addresses this by teaching:
identifying, based at least in part on the search, at least one evidence passage that serves as at least one potential link between the first concept and the second concept; (Stubley ¶ 0078-0081: the result of applying one or more search queries constructed by query builder 170 to unstructured data set(s) 140 may be a set of documents and/or passages identified as relevant to a search query and passed to evidence scorer 180. The set of returned documents/passages may be thresholded for relevance in any suitable way ... evaluate the retrieved passages in terms of the strength of supporting evidence that they provide for the extracted answer information as contributing to the best answer to the user's question)
analyzing, using natural language processing, the at least one evidence passage to determine that information included within the at least one evidence passage represents a connection between the first concept and the second concept; (Stubley FIG. 5, ¶ 0124: At act 530, documents containing natural language text may be analyzed, including analyzing one or more passages of text in the documents to determine whether the passage entails any of the hypotheses from the question. Exemplary techniques for entailment analysis are described above. At act 540, in response to determining that a passage entails a question hypothesis, the passage may be identified as providing supporting evidence for the generated answer to the question; Stubley ¶ 0084-0086: the strength of a passage's supporting evidence for answer information for a question portion may be evaluated by determining whether the passage entails a hypothesis corresponding to the question portion ... evidence scorer 180 may score passages based at least in part on the degree to which they entail one or more question hypotheses, and may rank passages based on their scores ... an entailment/contradiction score computed as described above ... may be used as one feature input to a statistical classifier used to score the strength of a passage's supporting evidence for an answer item)
Stubley further teaches generating a query result indicating the at least one evidence passage. (Stubley FIG. 5, ¶ 0123-0124: At act 550, the answer and the passage(s) identified as providing supporting evidence for that answer may be presented to the user in response to the input question. In some embodiments, as discussed above, multiple different passages may be scored based at least in part on the strength of the passages' supporting evidence for the answer to the question, and one or more of the passages may be selected for presentation to the user based on the passages' scores)
It would have been obvious to one of ordinary skill in the art before the effective filing date of the claimed invention to combine the functioning of the evidence passage retrieval of Tripathi with the evidence passage retrieval of Stubley.
In addition, both of the references (Tripathi and Stubley) disclose features that are directed to analogous art, and they are directed to the same field of endeavor, such as question answering interfaces.
Motivation to do so would be to improve the functioning of Tripathi returning search result documents in response to compound queries with the functioning in similar reference Stubley also returning search result documents in response to compound queries but with the improvement of scoring techniques.
Motivation to do so would also be the teaching, suggestion, or motivation for a person of ordinary skill in the art to implement designing a QA system to make intelligent use of both structured and unstructured data sources in its knowledge base as seen in Stubley ¶ 0028.
Tripathi in view of Stubley does not expressly disclose:
receiving, via a graphical user interface (GUI) presented via a user device, an input query that is associated with a research topic and that includes a first concept that serves as a starting point for the research topic and a second concept that serves as an intermediary point or an ending point of the research topic,
parsing, using a semantic parser, text of the input query to convert the text of the input query into a machine language that is a machine-understandable representation of a meaning of the text;
performing, based at least in part on the text of the input query in the machine language, a search of a plurality of knowledge data sources that each include a plurality of documents;
translating, using a natural language understanding (NLU) engine, the query result into a machine-readable structured representation of the query result; and
visually displaying, based at least in part on translating the query result and via the GUI presented via the user device, the query result.
However, Roberts taches:
receiving, via a graphical user interface (GUI) presented via a user device, an input query that is associated with a research topic and that includes a first concept that serves as a starting point for the research topic and a second concept that serves as an intermediary point or an ending point of the research topic, (Roberts FIG. 3, ¶ 0071: there is a hierarchy of information going from topics to entities to subtopics. As used herein, the term topic refers to any subject matter of interest, and the term “entity” refers to a particular type of topic having strongly typed attributes that distinguish it from other entities. As with topics, a subtopic does not imply any particular structure (although it may have structure), but it is expressly predicated on being about an entity; Roberts ¶ 0085-0097, see ¶ 0086: The first document 302 may generally include data relating to a topic identified by a topic identifier 310 ... the passage 312 may include a variety of textural references such as an entity identifier 314, an entity mention 316, and a citation 318; see ¶ 0087: An entity identifier 314 may include any mechanism for referring to a particular entity, e.g., an entity identified by the entity identifier 320 that is the focus of the entity profile 308 included in the second document 304; see ¶ 0093: the second document 304 may include a number of attributes 324, 328 and subtopics 340 used to describe the entity ... the attributes in the second document 304 may include one or more properties, traits, characteristics, details, features, relationships, and so forth, of an entity, and various schemas or folksonomies may capture such attributes 324 in a representation of the entity)
parsing, using a semantic parser, text of the input query to convert the text of the input query into a machine language that is a machine-understandable representation of a meaning of the text; (Roberts ¶ 0036: the server 106 may support search activity by periodically searching for content at remote locations on the data network 102 and indexing any resulting content for subsequent search by a client 104. This may include storing location or address information for a particular document as well as parsing the document in any suitable manner to identify words, images, media, metadata and the like, as well as the creation of feature vectors or other derivative data [relevant to machine-understandable representation] to assist in similarity-type comparisons, dissimilarity comparisons, or other analysis; show similarity-type comparisons used for querying in Roberts FIG. 6, ¶ 0159-0160: The extension may also show recommendations 614 from a recommendation engine that analyzes the documents bookmarked by the user and suggests other documents that are similar and may be helpful to the user ... the recommendation engine presents the recommendation 616 to the user; see machine-understandable representation of a meaning of the text through ¶ 0062-0066: techniques may be used to derive metadata that characterizes the content in more abstract forms. Thus for example the fifth data representation 226 may be a feature vector or feature collection constructed automatically from the semistructured output 206 of a tagger process 212. The purpose of a feature vector is to automatically capture aspects of a piece of unstructured or semistructured data in a form that is easily compared to other pieces of data... Feature vectors for text typically capture words and phrases and concepts derived from the words and phrases in a document)
performing, based at least in part on the text of the input query in the machine language, a search of a plurality of knowledge data sources that each include a plurality of documents; (Roberts ¶ 0036: the server 106 may support search activity by periodically searching for content at remote locations on the data network 102 [relevant to plurality of knowledge data sources] and indexing any resulting content for subsequent search by a client 104. This may include storing location or address information for a particular document as well as parsing the document in any suitable manner to identify words, images, media, metadata and the like, as well as the creation of feature vectors or other derivative data to assist in similarity-type comparisons, dissimilarity comparisons, or other analysis; Roberts FIG. 4, ¶ 0122-0123: The search results 418 may be located and retrieved for the target entity based on the entity profile 408 in the first window 402 using an algorithm 420 to traverse a data network or other source(s) of data ... The algorithm 420 may use the entity profile 408 as a query for retrieving, ranking, and displaying the search results 418. One possible approach to implementing the algorithm 420 is to construct a feature vector for the profile 408; see also Roberts ¶ 0005: the second window shows a plurality of search results from one or more sources that identify documents each containing a mention with a predetermined likelihood of referring to the entity)
translating, using a natural language understanding (NLU) engine, the query result into a machine-readable structured representation of the query result; and (Roberts FIG. 4, ¶ 0122: The second window 404 may be concurrently visible with the first window 402, and may display a plurality of search results 418 [shows a structured representation] … The search results 418 may be displayed with a preview of pertinent information included in the particular search result 418, such as highlighting or otherwise emphasizing keywords or other relevant text within a window of surrounding text [relevant to natural language understanding], where the keywords are specified by the user or determined by the algorithm 420; see more natural language understanding in at least Roberts ¶ 0088: The context surrounding an entity mention 316 may include a larger span of natural language prose near the entity mention 316 substring; see entity mentions in the query result in Roberts ¶ 0128: The third window 406 may include highlighting or other visual emphasis of mentions of the target entity 410 in the selected search result 428)
visually displaying, based at least in part on translating the query result and via the GUI presented via the user device, the query result. (Roberts FIG. 6, ¶ 0159-0160: The extension may also show recommendations 614 from a recommendation engine that analyzes the documents bookmarked by the user and suggests other documents that are similar and may be helpful to the user ... One of the recommendations 616 may be generated from analyzing a batch of documents related to the selected item in the foldering tree and finding common properties of those documents, such as long strings in common across the texts. After finding such a common property, the recommendation engine presents the recommendation 616 to the user; see also Roberts FIG. 4, ¶ 0116-0123, notably ¶ 0122: The second window 404 may be concurrently visible with the first window 402, and may display a plurality of search results 418 ... The search results 418 may be displayed with a preview of pertinent information included in the particular search result 418, such as highlighting or otherwise emphasizing keywords or other relevant text within a window of surrounding text, where the keywords are specified by the user or determined by the algorithm 420)
It would have been obvious to one of ordinary skill in the art before the effective filing date of the claimed invention to combine the functioning of the evidence passage retrieval of Tripathi as modified with the passage retrieval of Roberts.
In addition, both of the references (Tripathi as modified and Roberts) disclose features that are directed to analogous art, and they are directed to the same field of endeavor, such as textual data retrieval and display.
Motivation to do so would be to improve the functioning of Tripathi as modified returning search result documents in response to queries with the functioning in similar reference Roberts also returning search result documents in response to queries but with the improvement of in-progress entity profiles.
Motivation to do so would also be the teaching, suggestion, or motivation for a person of ordinary skill in the art to more quickly build a comprehensive description of an entity of interest as seen in Roberts ¶ 0004.
Regarding claim 15, Tripathi teaches:
A system comprising: memory; one or more processors; and one or more computer-executable instructions stored in the memory and executable by the one or more processors to perform operations comprising: (Tripathi ¶ 0002: the present disclosure also relates to computer readable medium containing program instructions for execution on a computer system, which when executed by a computer, cause the computer to perform method steps for presenting information related to search; Tripathi ¶ 0092: Examples of computing device include, but are not limited to, cellular phones, personal digital assistants (PDAs), handheld devices, wireless modems, laptop computers, personal computers, etc. Additionally, the computing device includes a casing, a memory, a processor)
receiving, via a graphical user interface (GUI) presented via a user device, an input query that is associated with a research topic and that includes a first concept … and a second concept … the research topic being associated with an inquiry regarding whether a relationship exists between the first concept and the second concept; (Tripathi FIG. 1, FIG. 5, ¶ 0124: At a step 502, a first input area, for receiving the search query from the user, is provided on a first user-interface of the computing device. At a step 504, one or more query segments and relations between the one or more query segments are displayed on a second area of the first user-interface. The received search query is developed to obtain the one or more query segments and the relations between the one or more query segments ... a list of one or more concepts [relevant to research topic] associated with the extracted search results)
performing, based at least in part on the text of the input query in the machine language, a search of a plurality of knowledge data sources that each include a plurality of documents; (Tripathi FIG. 1, FIG. 5, ¶ 0124: At a step 506, an arranged set of extracted search results is provided in a third area of the first user-interface. The search results are extracted, from at least one database [shows a plurality of knowledge data sources], based on the developed search query and arranged based on one or more parameters associated with the extracted search results ... the information related to the selected search result comprises a summary of the selected search result, one or more documents associated with the selected search result [shows plurality of documents])
…
generating a query result … that represents the connection between the first concept and the second concept; and (Tripathi FIG. 1, FIG. 5, ¶ 0124: The received search query is developed to obtain the one or more query segments and the relations between the one or more query segments [shows connection between concepts]. At a step 506, an arranged set of extracted search results is provided in a third area of the first user-interface. The search results are extracted, from at least one database, based on the developed search query [shows generating])
visually displaying, via the GUI presented via the user device, the query result. (Tripathi FIG. 1, FIG. 5, ¶ 0124: At step 510, information related to a selected search result or a selected concept, is provided on a second user-interface, in response to a selection input, from the user, based on one of a search result or a concept associated with the search results. Moreover, the information related to the selected search result comprises a summary of the selected search result, one or more documents associated with the selected search result, and concepts associated with the selected search result. Furthermore, the information related to the selected concept comprises a summary of the selected concept, and one or more documents associated with the selected concept)
Tripathi does not expressly disclose a first concept that serves as a starting point for the research topic and a second concept that serves as an intermediary point or an ending point of the research topic.
Tripathi further does not expressly disclose:
parsing, using a semantic parser, text of the input query to convert the text of the input query into a machine language that is a machine-understandable representation of a meaning of the text;
performing, based at least in part on the text of the input query in the machine language, a search of a plurality of knowledge data sources that each include a plurality of documents;
translating, using a natural language understanding (NLU) engine, the query result into a machine-readable structured representation of the query result; and
visually displaying, based at least in part on translating the query result and via the GUI presented via the user device, the query result.
identifying, based at least in part on the search, at least one evidence passage that serves as at least one potential link between the first concept and the second concept;
analyzing, using natural language processing, the at least one evidence passage to determine that information included within the at least one evidence passage represents a connection between the first concept and the second concept;
Tripathi further does not expressly disclose generating a query result indicating the at least one evidence passage.
However, Stubley addresses this by teaching:
identifying, based at least in part on the search, at least one evidence passage that serves as at least one potential link between the first concept and the second concept; (Stubley ¶ 0078-0081: the result of applying one or more search queries constructed by query builder 170 to unstructured data set(s) 140 may be a set of documents and/or passages identified as relevant to a search query and passed to evidence scorer 180. The set of returned documents/passages may be thresholded for relevance in any suitable way ... evaluate the retrieved passages in terms of the strength of supporting evidence that they provide for the extracted answer information as contributing to the best answer to the user's question)
analyzing, using natural language processing, the at least one evidence passage to determine that information included within the at least one evidence passage represents a connection between the first concept and the second concept; (Stubley FIG. 5, ¶ 0124: At act 530, documents containing natural language text may be analyzed, including analyzing one or more passages of text in the documents to determine whether the passage entails any of the hypotheses from the question. Exemplary techniques for entailment analysis are described above. At act 540, in response to determining that a passage entails a question hypothesis, the passage may be identified as providing supporting evidence for the generated answer to the question; Stubley ¶ 0084-0086: the strength of a passage's supporting evidence for answer information for a question portion may be evaluated by determining whether the passage entails a hypothesis corresponding to the question portion ... evidence scorer 180 may score passages based at least in part on the degree to which they entail one or more question hypotheses, and may rank passages based on their scores ... an entailment/contradiction score computed as described above ... may be used as one feature input to a statistical classifier used to score the strength of a passage's supporting evidence for an answer item)
Stubley further teaches generating a query result indicating the at least one evidence passage. (Stubley FIG. 5, ¶ 0123-0124: At act 550, the answer and the passage(s) identified as providing supporting evidence for that answer may be presented to the user in response to the input question. In some embodiments, as discussed above, multiple different passages may be scored based at least in part on the strength of the passages' supporting evidence for the answer to the question, and one or more of the passages may be selected for presentation to the user based on the passages' scores)
It would have been obvious to one of ordinary skill in the art before the effective filing date of the claimed invention to combine the functioning of the evidence passage retrieval of Tripathi with the evidence passage retrieval of Stubley.
In addition, both of the references (Tripathi and Stubley) disclose features that are directed to analogous art, and they are directed to the same field of endeavor, such as question answering interfaces.
Motivation to do so would be to improve the functioning of Tripathi returning search result documents in response to compound queries with the functioning in similar reference Stubley also returning search result documents in response to compound queries but with the improvement of scoring techniques.
Motivation to do so would also be the teaching, suggestion, or motivation for a person of ordinary skill in the art to implement designing a QA system to make intelligent use of both structured and unstructured data sources in its knowledge base as seen in Stubley ¶ 0028.
Tripathi in view of Stubley does not expressly disclose:
receiving, via a graphical user interface (GUI) presented via a user device, an input query that is associated with a research topic and that includes a first concept that serves as a starting point for the research topic and a second concept that serves as an intermediary point or an ending point of the research topic,
parsing, using a semantic parser, text of the input query to convert the text of the input query into a machine language that is a machine-understandable representation of a meaning of the text;
performing, based at least in part on the text of the input query in the machine language, a search of a plurality of knowledge data sources that each include a plurality of documents;
translating, using a natural language understanding (NLU) engine, the query result into a machine-readable structured representation of the query result; and
visually displaying, based at least in part on translating the query result and via the GUI presented via the user device, the query result.
However, Roberts taches:
receiving, via a graphical user interface (GUI) presented via a user device, an input query that is associated with a research topic and that includes a first concept that serves as a starting point for the research topic and a second concept that serves as an intermediary point or an ending point of the research topic, (Roberts FIG. 3, ¶ 0071: there is a hierarchy of information going from topics to entities to subtopics. As used herein, the term topic refers to any subject matter of interest, and the term “entity” refers to a particular type of topic having strongly typed attributes that distinguish it from other entities. As with topics, a subtopic does not imply any particular structure (although it may have structure), but it is expressly predicated on being about an entity; Roberts ¶ 0085-0097, see ¶ 0086: The first document 302 may generally include data relating to a topic identified by a topic identifier 310 ... the passage 312 may include a variety of textural references such as an entity identifier 314, an entity mention 316, and a citation 318; see ¶ 0087: An entity identifier 314 may include any mechanism for referring to a particular entity, e.g., an entity identified by the entity identifier 320 that is the focus of the entity profile 308 included in the second document 304; see ¶ 0093: the second document 304 may include a number of attributes 324, 328 and subtopics 340 used to describe the entity ... the attributes in the second document 304 may include one or more properties, traits, characteristics, details, features, relationships, and so forth, of an entity, and various schemas or folksonomies may capture such attributes 324 in a representation of the entity)
parsing, using a semantic parser, text of the input query to convert the text of the input query into a machine language that is a machine-understandable representation of a meaning of the text; (Roberts ¶ 0036: the server 106 may support search activity by periodically searching for content at remote locations on the data network 102 and indexing any resulting content for subsequent search by a client 104. This may include storing location or address information for a particular document as well as parsing the document in any suitable manner to identify words, images, media, metadata and the like, as well as the creation of feature vectors or other derivative data [relevant to machine-understandable representation] to assist in similarity-type comparisons, dissimilarity comparisons, or other analysis; show similarity-type comparisons used for querying in Roberts FIG. 6, ¶ 0159-0160: The extension may also show recommendations 614 from a recommendation engine that analyzes the documents bookmarked by the user and suggests other documents that are similar and may be helpful to the user ... the recommendation engine presents the recommendation 616 to the user; see machine-understandable representation of a meaning of the text through ¶ 0062-0066: techniques may be used to derive metadata that characterizes the content in more abstract forms. Thus for example the fifth data representation 226 may be a feature vector or feature collection constructed automatically from the semistructured output 206 of a tagger process 212. The purpose of a feature vector is to automatically capture aspects of a piece of unstructured or semistructured data in a form that is easily compared to other pieces of data... Feature vectors for text typically capture words and phrases and concepts derived from the words and phrases in a document)
performing, based at least in part on the text of the input query in the machine language, a search of a plurality of knowledge data sources that each include a plurality of documents; (Roberts ¶ 0036: the server 106 may support search activity by periodically searching for content at remote locations on the data network 102 [relevant to plurality of knowledge data sources] and indexing any resulting content for subsequent search by a client 104. This may include storing location or address information for a particular document as well as parsing the document in any suitable manner to identify words, images, media, metadata and the like, as well as the creation of feature vectors or other derivative data to assist in similarity-type comparisons, dissimilarity comparisons, or other analysis; Roberts FIG. 4, ¶ 0122-0123: The search results 418 may be located and retrieved for the target entity based on the entity profile 408 in the first window 402 using an algorithm 420 to traverse a data network or other source(s) of data ... The algorithm 420 may use the entity profile 408 as a query for retrieving, ranking, and displaying the search results 418. One possible approach to implementing the algorithm 420 is to construct a feature vector for the profile 408; see also Roberts ¶ 0005: the second window shows a plurality of search results from one or more sources that identify documents each containing a mention with a predetermined likelihood of referring to the entity)
translating, using a natural language understanding (NLU) engine, the query result into a machine-readable structured representation of the query result; and (Roberts FIG. 4, ¶ 0122: The second window 404 may be concurrently visible with the first window 402, and may display a plurality of search results 418 [shows a structured representation] … The search results 418 may be displayed with a preview of pertinent information included in the particular search result 418, such as highlighting or otherwise emphasizing keywords or other relevant text within a window of surrounding text [relevant to natural language understanding], where the keywords are specified by the user or determined by the algorithm 420; see more natural language understanding in at least Roberts ¶ 0088: The context surrounding an entity mention 316 may include a larger span of natural language prose near the entity mention 316 substring; see entity mentions in the query result in Roberts ¶ 0128: The third window 406 may include highlighting or other visual emphasis of mentions of the target entity 410 in the selected search result 428)
visually displaying, based at least in part on translating the query result and via the GUI presented via the user device, the query result. (Roberts FIG. 6, ¶ 0159-0160: The extension may also show recommendations 614 from a recommendation engine that analyzes the documents bookmarked by the user and suggests other documents that are similar and may be helpful to the user ... One of the recommendations 616 may be generated from analyzing a batch of documents related to the selected item in the foldering tree and finding common properties of those documents, such as long strings in common across the texts. After finding such a common property, the recommendation engine presents the recommendation 616 to the user; see also Roberts FIG. 4, ¶ 0116-0123, notably ¶ 0122: The second window 404 may be concurrently visible with the first window 402, and may display a plurality of search results 418 ... The search results 418 may be displayed with a preview of pertinent information included in the particular search result 418, such as highlighting or otherwise emphasizing keywords or other relevant text within a window of surrounding text, where the keywords are specified by the user or determined by the algorithm 420)
It would have been obvious to one of ordinary skill in the art before the effective filing date of the claimed invention to combine the functioning of the evidence passage retrieval of Tripathi as modified with the passage retrieval of Roberts.
In addition, both of the references (Tripathi as modified and Roberts) disclose features that are directed to analogous art, and they are directed to the same field of endeavor, such as textual data retrieval and display.
Motivation to do so would be to improve the functioning of Tripathi as modified returning search result documents in response to queries with the functioning in similar reference Roberts also returning search result documents in response to queries but with the improvement of in-progress entity profiles.
Motivation to do so would also be the teaching, suggestion, or motivation for a person of ordinary skill in the art to more quickly build a comprehensive description of an entity of interest as seen in Roberts ¶ 0004.
Regarding claims 2, 9, and 16, Tripathi in view of Stubley and Roberts teaches:
wherein a potential link of the at least one potential link is a structured relational representation that connects the first concept and the second concept, (Stubley ¶ 0078-0081, see first ¶ 0078: evaluate the retrieved passages in terms of the strength of supporting evidence that they provide for the extracted answer information as contributing to the best answer to the user's question; Stubley ¶ 0079: Evidence scorer 180 may extract answer information from retrieved relevant documents/passages in any suitable way ... extract one or more assertions from a passage to match one or more intent and/or constraint relations from the user's question; Stubley ¶ 0081: evidence scorer 180 may extract the assertion, “Avatar has tall blue aliens,” as being made by the passage. This assertion matches the question constraint “movie X has tall blue aliens” with a specific concept (“Avatar”) replacing the generic/unknown concept (“movie X”) in the constraint, such that the assertion provides candidate answer information for the question constraint)
and wherein the at least one evidence passage includes one or more portions of a knowledge data source of the plurality of knowledge data sources. (Stubley FIG. 5, ¶ 0120-0122: the structured data source(s) may include one or more databases, and in some embodiments, the unstructured data source(s) may include one or more sets of documents containing natural language text [shows plurality of knowledge data sources] … At act 430, one or more first queries may be constructed from the first question portion(s) answerable from the structured data source(s), and may be applied to the structured data source(s) to retrieve first answer information for the first question portion(s) Likewise, one or more second queries may be constructed from the second question portion(s) answerable from the unstructured data source(s), and may be applied to the unstructured data source(s) to retrieve second answer information for the second question portion(s) ... At act 440, answer information from the structured and unstructured data sources may be merged to form an answer to the user's question, and this answer may be presented to the user at act 450. In some embodiments, as discussed above, one or more portions of natural language text from the unstructured data source(s) may be identified as providing evidence that supports answer information retrieved from the unstructured data source(s), and this natural language text (e.g., one or more supporting passages) may be presented to the user in association with the generated answer to the user's question)
Regarding claims 11 and 18, Tripathi in view of Stubley and Roberts teaches:
wherein the query result includes a portion of the at least one evidence passage that provides evidentiary support for the connection between the first concept and the second concept. (Stubley FIG. 5, ¶ 0123-0124: At act 550, the answer and the passage(s) identified as providing supporting evidence for that answer may be presented to the user in response to the input question. In some embodiments, as discussed above, multiple different passages may be scored based at least in part on the strength of the passages' supporting evidence for the answer to the question, and one or more of the passages may be selected for presentation to the user based on the passages' scores; see connections between concepts in Stubley ¶ 0081: evidence scorer 180 may extract the assertion, “Avatar has tall blue aliens,” as being made by the passage. This assertion matches the question constraint “movie X has tall blue aliens” with a specific concept (“Avatar”) replacing the generic/unknown concept (“movie X”) in the constraint, such that the assertion provides candidate answer information for the question constraint)
Regarding claims 5, 12, and 19 , Tripathi in view of Stubley and Roberts teaches:
determining a first score indicating a first strength of the connection between the first concept and the second concept; (Stubley ¶ 0082: evidence scorer 180 may evaluate passages in terms of the strength of the evidence they provide in support of the extracted answer information as contributing to accurately answering the user's question; Stubley ¶ 0085-0086: evidence scorer 180 may score passages based at least in part on the degree to which they entail one or more question hypotheses, and may rank passages based on their scores)
identifying, based at least in part on the search, at least one second evidence passage that serves as at least one second potential link between the first concept and the second concept; (Stubley ¶ 0078-0083; see first Stubley ¶ 0078: the result of applying one or more search queries constructed by query builder 170 to unstructured data set(s) 140 may be a set of documents and/or passages [shows at least one second evidence passage] identified as relevant to a search query and passed to evidence scorer 180. The set of returned documents/passages may be thresholded for relevance in any suitable way ... evaluate the retrieved passages in terms of the strength of supporting evidence that they provide for the extracted answer information as contributing to the best answer to the user's question; see then Stubley ¶ 0083: evidence scorer 180 (or any other suitable component of QA system 100, such as question analyzer 160 or query builder 170) may map the user's question to one or more hypotheses to be tested for entailment against one or more assertions extracted from natural language text passages)
analyzing, using the natural language processing, the at least one second evidence passage to determine that second information included within the at least one second evidence passage represents the connection between the first concept and the second concept; and (Stubley FIG. 5, ¶ 0124: At act 530, documents containing natural language text may be analyzed, including analyzing one or more passages of text in the documents [shows natural language processing] to determine whether the passage entails any of the hypotheses from the question. Exemplary techniques for entailment analysis are described above. At act 540, in response to determining that a passage entails a question hypothesis, the passage may be identified as providing supporting evidence for the generated answer to the question)
including the at least one second evidence passage in the query result. (Stubley FIG. 5, ¶ 0123-0124: At act 550, the answer and the passage(s) [shows at least one second evidence passage] identified as providing supporting evidence for that answer may be presented to the user in response to the input question. In some embodiments, as discussed above, multiple different passages may be scored based at least in part on the strength of the passages' supporting evidence for the answer to the question, and one or more of the passages may be selected for presentation to the user based on the passages' scores)
Regarding claims 6 and 13, Tripathi in view of Stubley and Roberts teaches:
wherein the at least one potential link includes at least one relational representation that connects the first concept and the second concept, (Stubley ¶ 0078-0081; see first Stubley ¶ 0078: the result of applying one or more search queries constructed by query builder 170 to unstructured data set(s) 140 may be a set of documents and/or passages identified as relevant to a search query and passed to evidence scorer 180. The set of returned documents/passages may be thresholded for relevance in any suitable way ... evaluate the retrieved passages in terms of the strength of supporting evidence that they provide for the extracted answer information as contributing to the best answer to the user's question; see then Stubley ¶ 0081: evidence scorer 180 may extract the assertion, “Avatar has tall blue aliens,” as being made by the passage. This assertion matches the question constraint “movie X has tall blue aliens” with a specific concept (“Avatar”) replacing the generic/unknown concept (“movie X”) in the constraint, such that the assertion provides candidate answer information for the question constraint)
further comprising determining at least one relation cluster by aggregating the at least one relational representation based at least in part on a degree of semantic similarity between the at least one relational representation. (Stubley ¶ 0079: evidence scorer 180 may use concept ID and/or semantic parser annotations from retrieved indexed passages, together with any domain model knowledge about the particular relevance of corresponding section headers, document titles, etc., to extract one or more assertions from a passage to match one or more intent and/or constraint relations from the user's question; Stubley ¶ 0086-0104: Any other suitable features may be used in such a classifier, in addition to or instead of the above entailment/contradiction score feature, some non-limiting examples of which may include: ... Brown Clustering PredArg Matcher: Returns an average of the distance among the terms in a semantic relation match. For example, if there is an exact match between the first arguments of a relation (0) but no match between the second (1), the relation's score may be 0.5)
Claims 4, 7; 10, 14; 17, and 20-21 are rejected under 35 U.S.C. 103 as being unpatentable over Tripathi in view of Stubley and Roberts in further view of Bagchi et al., U.S. Patent Application Publication No. 2018/0025127 (hereinafter Bagchi).
Regarding claim 4, Tripathi in view of Stubley and Roberts teaches all the features with respect to claim 1 above including:
wherein: the query result includes a portion of the at least one evidence passage that provides evidentiary support for the connection between the first concept and the second concept. (Stubley FIG. 5, ¶ 0123-0124: At act 550, the answer and the passage(s) identified as providing supporting evidence for that answer may be presented to the user in response to the input question. In some embodiments, as discussed above, multiple different passages may be scored based at least in part on the strength of the passages' supporting evidence for the answer to the question, and one or more of the passages may be selected for presentation to the user based on the passages' scores; see connections between concepts in Stubley ¶ 0081: evidence scorer 180 may extract the assertion, “Avatar has tall blue aliens,” as being made by the passage. This assertion matches the question constraint “movie X has tall blue aliens” with a specific concept (“Avatar”) replacing the generic/unknown concept (“movie X”) in the constraint, such that the assertion provides candidate answer information for the question constraint)
Tripathi in view of Stubley and Roberts does not expressly disclose:
the at least one evidence passage indicates that the first concept causes, or induces, the second concept; and
However, Bagchi addresses this by teaching:
the at least one evidence passage indicates that the first concept causes, or induces, the second concept. (Bagchi ¶ 0041: The medical examples found herein illustrates this through answers, confidences, dimensions of evidence, associated evidence passages, and documents where this evidence is found, as well as, reliability of the evidence source ... Examples of queries include (but are not limited to): what clinical conditions are characterized by a set of symptoms?; what is the “differential diagnosis” (a ranked list of diseases) that could potentially cause a set of symptoms, conditions, findings?; Bagchi ¶ 0059-0060: a clinical diagnosis query containing symptoms, findings, family history and demographic information, could generate a series of queries as follows, where the text in the <>characters is replaced by the corresponding concepts found in the case text: “What disease of condition could cause <symptom>?”; “What disease of condition could cause <symptom>and <findings>?; “What disease of condition could cause <symptom>, <findings>and <family history>?; “What disease of condition could cause <symptom>, <findings>, <family history>and <demographics>?; etc. ... The method receives answers from the question-answering system in item 210. For each query submitted, the question-answering system 110 returns a list of answers, their confidences, evidence dimensions, and evidence sources)
It would have been obvious to one of ordinary skill in the art before the effective filing date of the claimed invention to combine the functioning of the evidence scoring of Tripathi as modified with the evidence scoring of Bagchi.
In addition, both of the references (Tripathi as modified and Bagchi) disclose features that are directed to analogous art, and they are directed to the same field of endeavor, such as question answering interfaces.
Motivation to do so would also be the teaching, suggestion, or motivation for a person of ordinary skill in the art to improve hypotheses as seen in Bagchi ¶ 0011.
Regarding claims 10 and 17, Tripathi in view of Stubley and Roberts teaches all the features with respect to claims 8 and 15 above respectively but does not expressly disclose:
wherein the at least one evidence passage indicates that the first concept causes, or induces, the second concept.
However, Bagchi addresses this by teaching:
wherein the at least one evidence passage indicates that the first concept causes, or induces, the second concept. (Bagchi ¶ 0041: The medical examples found herein illustrates this through answers, confidences, dimensions of evidence, associated evidence passages, and documents where this evidence is found, as well as, reliability of the evidence source ... Examples of queries include (but are not limited to): what clinical conditions are characterized by a set of symptoms?; what is the “differential diagnosis” (a ranked list of diseases) that could potentially cause a set of symptoms, conditions, findings?; Bagchi ¶ 0059-0060: a clinical diagnosis query containing symptoms, findings, family history and demographic information, could generate a series of queries as follows, where the text in the <>characters is replaced by the corresponding concepts found in the case text: “What disease of condition could cause <symptom>?”; “What disease of condition could cause <symptom>and <findings>?; “What disease of condition could cause <symptom>, <findings>and <family history>?; “What disease of condition could cause <symptom>, <findings>, <family history>and <demographics>?; etc. ... The method receives answers from the question-answering system in item 210. For each query submitted, the question-answering system 110 returns a list of answers, their confidences, evidence dimensions, and evidence sources)
It would have been obvious to one of ordinary skill in the art before the effective filing date of the claimed invention to combine the functioning of the evidence scoring of Tripathi as modified with the evidence scoring of Bagchi.
In addition, both of the references (Tripathi as modified and Bagchi) disclose features that are directed to analogous art, and they are directed to the same field of endeavor, such as question answering interfaces.
Motivation to do so would also be the teaching, suggestion, or motivation for a person of ordinary skill in the art to improve hypotheses as seen in Bagchi ¶ 0011.
Regarding claims 7 and 14, Tripathi in view of Stubley and Roberts teaches:
determining an aggregation confidence associated with a relation cluster of the at least one relation cluster, the aggregation confidence being based at least in part on a … score of a portion of the at least one evidence passage. (Stubley ¶ 0109: answer items may be scored based on the evidence scores (e.g., entailment confidence levels) of their respective supporting passages. In some embodiments, the score of an answer item may depend at least in part on how many different passages or documents support that answer item)
Tripathi in view of Stubley and Roberts does not expressly disclose a reliability score, taught by Bagchi. (Bagchi ¶ 0041: The medical examples found herein illustrates this through answers, confidences, dimensions of evidence, associated evidence passages, and documents where this evidence is found, as well as, reliability of the evidence source)
It would have been obvious to one of ordinary skill in the art before the effective filing date of the claimed invention to combine the functioning of the evidence scoring of Tripathi as modified with the evidence scoring of Bagchi.
In addition, both of the references (Tripathi as modified and Bagchi) disclose features that are directed to analogous art, and they are directed to the same field of endeavor, such as question answering interfaces.
Motivation to do so would be to improve the functioning of Tripathi as modified scoring answers with the functioning in similar reference Bagchi also scoring answers but with the improvement of a variety of evidence techniques.
Regarding claim 20, Tripathi in view of Stubley and Roberts teaches all the features with respect to claim 15 above including:
wherein the at least one potential link includes at least one relational representation that connects the first concept and the second concept, (Stubley ¶ 0078-0081; see first Stubley ¶ 0078: the result of applying one or more search queries constructed by query builder 170 to unstructured data set(s) 140 may be a set of documents and/or passages identified as relevant to a search query and passed to evidence scorer 180. The set of returned documents/passages may be thresholded for relevance in any suitable way ... evaluate the retrieved passages in terms of the strength of supporting evidence that they provide for the extracted answer information as contributing to the best answer to the user's question; see then Stubley ¶ 0081: evidence scorer 180 may extract the assertion, “Avatar has tall blue aliens,” as being made by the passage. This assertion matches the question constraint “movie X has tall blue aliens” with a specific concept (“Avatar”) replacing the generic/unknown concept (“movie X”) in the constraint, such that the assertion provides candidate answer information for the question constraint)
wherein the operations further comprise: determining at least one relation cluster by aggregating the at least one relational representation based at least in part on a degree of semantic similarity between the at least one relational representation; and (Stubley ¶ 0079: evidence scorer 180 may use concept ID and/or semantic parser annotations from retrieved indexed passages, together with any domain model knowledge about the particular relevance of corresponding section headers, document titles, etc., to extract one or more assertions from a passage to match one or more intent and/or constraint relations from the user's question; Stubley ¶ 0086-0104: Any other suitable features may be used in such a classifier, in addition to or instead of the above entailment/contradiction score feature, some non-limiting examples of which may include: ... Brown Clustering PredArg Matcher: Returns an average of the distance among the terms in a semantic relation match. For example, if there is an exact match between the first arguments of a relation (0) but no match between the second (1), the relation's score may be 0.5)
determining an aggregation confidence associated with a relation cluster of the at least one relation cluster, the aggregation confidence being based at least in part on a … score of a portion of the at least one evidence passage. (Stubley ¶ 0109: answer items may be scored based on the evidence scores (e.g., entailment confidence levels) of their respective supporting passages. In some embodiments, the score of an answer item may depend at least in part on how many different passages or documents support that answer item)
Tripathi in view of Stubley does not expressly disclose a reliability score, taught by Bagchi. (Bagchi ¶ 0041: The medical examples found herein illustrates this through answers, confidences, dimensions of evidence, associated evidence passages, and documents where this evidence is found, as well as, reliability of the evidence source)
It would have been obvious to one of ordinary skill in the art before the effective filing date of the claimed invention to combine the functioning of the evidence scoring of Tripathi as modified with the evidence scoring of Bagchi.
In addition, both of the references (Tripathi as modified and Bagchi) disclose features that are directed to analogous art, and they are directed to the same field of endeavor, such as question answering interfaces.
Motivation to do so would be to improve the functioning of Tripathi as modified scoring answers with the functioning in similar reference Bagchi also scoring answers but with the improvement of a variety of evidence techniques.
Regarding claim 21, Tripathi in view of Stubley and Roberts teaches all the features with respect to claim 1 above including generating … a multi-dimensional … query result. (Roberts FIG. 5, ¶ 0156-0157: the two-dimensional slider 502 may act as a thresholding tool. By selecting a location in the two-dimensional slider 502, the user can restrict the search results that are returned by a search algorithm to results that have relevance and confidence scores above the selected thresholds)
Tripathi in view of Stubley and Roberts does not expressly disclose:
wherein translating the query result comprises generating, by the NLU engine, a multi-dimensional interpretation of the query result.
However, Bagchi addresses this by teaching:
wherein translating the query result comprises generating, by the NLU engine, a multi-dimensional interpretation of the query result. (Bagchi FIG. 2, FIG. 4, ¶ 0063-0064: In item 214, the method displays information to support decision-making. The list of answers is displayed along with answer confidences for the decision-maker 108 to evaluate (see FIG. 4 for an example). Thus, the method outputs the queries, the answers, the corresponding confidence values, the links to the evidence sources, and the numerical value of each evidence dimension using the input/output module upon user inquiry ... embodiments herein automatically and continuously update the diagnosis answers, the corresponding confidence values, and the numerical value of each evidence dimension based on revisions to the problem case information to produce revised queries, answers, corresponding confidence values, etc. (using the question-answering module). This method can also automatically output the revised queries, answers, and/or corresponding confidence values when a difference threshold is exceeded. This “difference threshold” can comprise a time period (e.g., hours, weeks, months, etc.), the amount one or more answers change (e.g., percentage change, polarity (yes/no) change, number of answers changing, etc.) and/or an amount of confidence value changes (percent confidence change, confidence polarity change, etc.))
It would have been obvious to one of ordinary skill in the art before the effective filing date of the claimed invention to combine the functioning of the evidence scoring of Tripathi as modified with the evidence scoring of Bagchi.
In addition, both of the references (Tripathi as modified and Bagchi) disclose features that are directed to analogous art, and they are directed to the same field of endeavor, such as question answering interfaces.
Motivation to do so would be to improve the functioning of Tripathi as modified scoring answers with the functioning in similar reference Bagchi also scoring answers but with the improvement of a variety of evidence techniques.
Conclusion
The prior art made of record and not relied upon is considered pertinent to applicant's disclosure.
Walsh, U.S. Patent Application Publication No. 2012/0204104; see Walsh FIGs. 2, ¶ 0043-0045 reciting a document analysis graphical user interface (GUI) allowing a researcher to start a research project by entering one or more concepts (with corresponding words/word groups), relevant to at least the independent claim limitations involving an input query that is associated with a research topic and that includes a first concept that serves as a starting point for the research topic.
Applicant's amendment necessitated the new ground(s) of rejection presented in this Office action. Accordingly, THIS ACTION IS MADE FINAL. See MPEP § 706.07(a). 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.
Any inquiry concerning this communication or earlier communications from the examiner should be directed to JEDIDIAH P FERRER whose telephone number is (571)270-7695. The examiner can normally be reached Monday, Tuesday, Friday, 12:00pm-9:00pm.
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.
/J.P.F/Examiner, Art Unit 2153 July 21, 2026
/KAVITA STANLEY/Supervisory Patent Examiner, Art Unit 2153
Prosecution Insights