# US 11,443,213 B2 · System and method for approximate reasoning using ontologies and unstructured data > Agents: this data is available as JSON from the patentgrep API and MCP server — see https://patentgrep.com/llms.txt - Number: US 11,443,213 B2 - Document ID: US11443213B2 - Status: Granted - Kind: B2 - Application number: 16/117,992 - Filed: 2018-08-30 - Published: 2020-03-05 - Granted: 2022-09-13 - Earliest claimed: 2018-08-30 - Assignees: International Business Machines Corporation - Applicants: International Business Machines Corporation - Inventors: Hima Prasad Karanam; Shajith Ikbal Mohamed; Sumit Bhatia; Sumit Neelam; L. Venkata Subramaniam; Udit Sharma - Web page: https://patentgrep.com/patent/US11443213B2 ## CPC - G06F 16/367: {Ontology} - G06N 5/048: Fuzzy inferencing - G06F 16/334: Query execution (filtering based on additional data G06F16/335) - G06N 5/022: Knowledge engineering; Knowledge acquisition - G06N 5/04: Inference or reasoning models - G06N 20/00: Machine learning - G06F: ELECTRIC DIGITAL DATA PROCESSING (computer systems based on specific computational models G06N) - G06N: COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS ## Abstract Methods, systems and computer program products for query processing are provided herein. A computer-implemented method includes receiving a first query from a user, determining whether the first query is capable of being answered using symbolic reasoning performed on data of a symbolic knowledge base, and executing the symbolic reasoning to generate a first query answer in response to a determination that the first query is capable of being answered using the symbolic reasoning. Axioms are extracted from a plurality of documents when it is determined that a second query is not capable of being answered using the symbolic reasoning. The method further includes determining whether the axioms are consistent with the symbolic knowledge base, and generating a second query answer based on the axioms in response to a determination that the one or more axioms are consistent with the symbolic knowledge base. ## Claims ### Claim 1 A computer-implemented method, comprising: receiving a first query from at least one user, wherein the first query comprises a natural language format query; converting the first query from the natural language format to a SPARQL format determining whether the first query is capable of being answered using symbolic reasoning performed on data of a symbolic knowledge base; executing the symbolic reasoning to generate a first query answer in response to a determination that the first query is capable of being answered using the symbolic reasoning, wherein executing the symbolic reasoning comprises automatically processing data from at least a portion of the symbolic knowledge base using a set of one or more artificial intelligence algorithms and one or more portions of data derived from the first query; receiving a second query from the at least one user, wherein the second query comprises a natural language format query; converting the second query from the natural language format to a SPARQL format; extracting one or more axioms from a plurality of documents in response to a determination that a second query is not capable of being answered using the symbolic reasoning; determining whether the one or more axioms are consistent with the symbolic knowledge base; and generating a second query answer based on the one or more axioms in response to a determination that the one or more axioms are consistent with the symbolic knowledge base. ### Claim 2 (depends on 1) The computer-implemented method of claim 1, comprising: determining whether part of a given query is capable of being answered using the symbolic reasoning in response to the determination that the given query is not capable of being answered. ### Claim 3 (depends on 1) The computer-implemented method of claim 1, comprising: modifying a given query to at least one of (i) include one or more keywords and (ii) exclude one or more keywords in response to the determination that the given query is not capable of being answered using the symbolic reasoning. ### Claim 4 (depends on 3) The computer-implemented method of claim 3, wherein the one or more keywords correspond to one or more elements of an ontology. ### Claim 5 (depends on 1) The computer-implemented method of claim 1, wherein the one or more axioms comprise at least one of (i) an assertion component and (ii) a terminological component. ### Claim 6 (depends on 1) The computer-implemented method of claim 1, comprising: adding the one or more axioms determined to be consistent to the symbolic knowledge base. ### Claim 7 (depends on 1) The computer-implemented method of claim 1, wherein the plurality of documents comprise unstructured textual documents. ### Claim 8 (depends on 1) The computer-implemented method of claim 1, wherein the symbolic knowledge base comprises structured data represented in one or more ontologies. ### Claim 9 (depends on 1) The computer-implemented method of claim 1, wherein said generating the second query answer comprises performing the symbolic reasoning in connection with at least part of the query. ### Claim 10 (depends on 1) The computer-implemented method of claim 1, wherein said executing the symbolic reasoning comprises (i) identifying a portion of an ontology in the symbolic knowledge base required for generating the first query answer and (ii) generating the first query answer. ### Claim 11 (depends on 1) The computer-implemented method of claim 1, wherein said extracting is driven by one or more relations in one or more ontologies in the symbolic knowledge base. ### Claim 12 (depends on 1) The computer implemented method of claim 1, comprising: generating a third query answer in response to a determination that the one or more axioms are inconsistent with the symbolic knowledge base, wherein the third query answer comprises an answer based on (i) structured data from the symbolic knowledge base and (ii) unstructured data from the plurality of documents. ### Claim 13 A computer program product, the computer program product comprising a computer readable storage medium having program instructions embodied therewith, the program instructions executable by a device to cause the device to: receive a first query from at least one user, wherein the first query comprises a natural language format query; convert the first query from the natural language format to a SPARQL format; determine whether the first query is capable of being answered using symbolic reasoning performed on data of a symbolic knowledge base; execute the symbolic reasoning to generate a first query answer in response to a determination that the first query is capable of being answered using the symbolic reasoning, wherein executing the symbolic reasoning comprises automatically processing data from at least a portion of the symbolic knowledge base using a set of one or more artificial intelligence algorithms and one or more portions of data derived from the first query; receive a second query from the at least one user, wherein the second query comprises a natural language format query; convert the second query from the natural language format to a SPARQL format; extract one or more axioms from a plurality of documents in response to a determination that a second query is not capable of being answered using the symbolic reasoning; determine whether the one or more axioms are consistent with the symbolic knowledge base; and generate a second query answer based on the one or more axioms in response to a determination that the one or more axioms are consistent with the symbolic knowledge base. ### Claim 14 A system comprising: a memory; and at least one processor coupled to the memory and configured for: receiving a first query from at least one user, wherein the first query comprises a natural language format query; converting the first query from the natural language format to a SPARQL format; determining whether the first query is capable of being answered using symbolic reasoning performed on data of a symbolic knowledge base; executing the symbolic reasoning to generate a first query answer in response to a determination that the first query is capable of being answered using the symbolic reasoning, wherein executing the symbolic reasoning comprises automatically processing data from at least a portion of the symbolic knowledge base using a set of one or more artificial intelligence algorithms and one or more portions of data derived from the first query; receiving a second query from the at least one user, wherein the second query comprises a natural language format query; converting the second query from the natural language format to a SPARQL format; extracting one or more axioms from a plurality of documents in response to a determination that a second query is not capable of being answered using the symbolic reasoning; determining whether the one or more axioms are consistent with the symbolic knowledge base; and generating a second query answer based on the one or more axioms in response to a determination that the one or more axioms are consistent with the symbolic knowledge base. ### Claim 15 (depends on 14) The system according to claim 14, wherein the at least one processor is further configured for: determining whether part of a given query is capable of being answered using the symbolic reasoning in response to the determination that the given query is not capable of being answered.