# US 11,030,260 B2 · System and method for dynamically normalized semantic distance and applications thereof > Agents: this data is available as JSON from the patentgrep API and MCP server — see https://patentgrep.com/llms.txt - Number: US 11,030,260 B2 - Document ID: US11030260B2 - Status: Granted - Kind: B2 - Application number: 16/116,007 - Filed: 2018-08-29 - Published: 2020-03-05 - Granted: 2021-06-08 - Earliest claimed: 2018-08-29 - Assignees: IP.com I, LLC - Applicants: IP.com I, LLC - Inventors: Wen Ruan; Samuel C. Baxter; James Thomas Durkin; William Yurich Fowlkes - Web page: https://patentgrep.com/patent/US11030260B2 ## CPC - G06F 16/3332: Query translation - G06F 16/93: Document management systems - G06F 16/22: Indexing; Data structures therefor; Storage structures - G06F 16/245: Query processing - G06F 16/24578: {using ranking} - G06F 16/334: Query execution (filtering based on additional data G06F16/335) - G06F: ELECTRIC DIGITAL DATA PROCESSING (computer systems based on specific computational models G06N) ## Abstract The present teaching relates to method and system for assessing semantic distance. A query is received and processed, by a document representation constructor, to obtain a first query representation. The query is then processed by a query representation constructor to obtain a second query representation. Based on the first and second query representations, a dynamic norm is determined. Based on the second query representation, a plurality of documents are identified as being related to the query and a semantic distance between the query and the plurality of documents is determined based on the dynamic norm. ## Claims ### Claim 1 A method, implemented on a machine having at least one processor, storage, and a communication platform for assessing semantic distance, comprising: receiving, via the communication platform, a query; obtaining, via a document representation constructor, a first query representation of the query capturing characteristics of the query; obtaining, via a query representation constructor, a second query representation of the query capturing the characteristics of the query; determining, via a dynamic norm determiner, a dynamic norm based on a discrepancy between the first and the second query representations, wherein the dynamic norm is indicative of a bias between the document and query representation constructors; identifying a plurality of documents relevant to the query based on the second query representation; and for each of the plurality of documents, computing a similarity between the document represented by a document representation obtained via the document representation constructor and the query represented by the second query representation obtained via the query representation constructor, and determining a semantic distance between the query and the document based on the dynamic norm and the similarity, wherein the semantic distance is derived by adjusting a distance between the query and the document based on the dynamic norm to address the bias. ### Claim 2 (depends on 1) The method of claim 1, wherein the determining the dynamic norm comprises: computing a similarity between the first and the second query representations; and determining the dynamic norm based on the similarity. ### Claim 3 (depends on 1) The method of claim 1, wherein the step of the determining the semantic distance comprises: computing a similarity between the query and each of the plurality of relevant documents; determining a statistic based on the similarities between the query and the plurality of relevant documents; and obtaining the semantic distance based on the statistic and the dynamic norm. ### Claim 4 (depends on 1) The method of claim 1, wherein the semantic distance is determined further based on a confidence score, which is estimated based on one or more parameters characterizing a distribution pattern of a plurality of similarities measured between the query and the plurality of relevant documents, respectively. ### Claim 5 (depends on 4) The method of claim 4, wherein the one or more parameters characterize a curve formed by connecting the plurality of similarities. ### Claim 6 (depends on 1) The method of claim 1, further comprising computing at least one of: a novelty score characterizing novelty of the query with respect to the plurality of relevant documents; and a non-obviousness score characterizing non-obviousness of the query with respect to the plurality of relevant documents. ### Claim 7 (depends on 6) The method of claim 6, further comprising: computing a distinction index of the query based on the novelty score and/or the non-obviousness score. ### Claim 8 A machine readable non-transitory medium having information recorded thereon for assessing semantic distance, wherein the information, when read by the machine, causes the machine to perform the following: receiving, via the communication platform, a query; obtaining, via a document representation constructor, a first query representation of the query capturing characteristics of the query; obtaining, via a query representation constructor, a second query representation of the query capturing the characteristics of the query; determining, via a dynamic norm determiner, a dynamic norm based on a discrepancy between the first and the second query representations, wherein the dynamic norm is indicative of a bias between the document and query representation constructors; identifying a plurality of documents relevant to the query based on the second query representation; and for each of the plurality of documents, computing a similarity between the document represented by a document representation obtained via the document representation constructor and the query represented by the second query representation obtained via the query representation constructor, and determining a semantic distance between the query and the document based on the dynamic norm and the similarity, wherein the semantic distance is derived by adjusting a distance between the query and the document based on the dynamic norm to address the bias. ### Claim 9 (depends on 8) The medium of claim 8, wherein the determining the dynamic norm comprises: computing a similarity between the first and the second query representations; and determining the dynamic norm based on the similarity. ### Claim 10 (depends on 8) The medium of claim 8, wherein the step of the determining the semantic distance comprises: computing a similarity between the query and each of the plurality of relevant documents; determining a statistic based on the similarities between the query and the plurality of relevant documents; and obtaining the semantic distance based on the statistic and the dynamic norm. ### Claim 11 (depends on 8) The medium of claim 8, wherein the semantic distance is determined further based on a confidence score, which is estimated based on one or more parameters characterizing a distribution pattern of a plurality of similarities measured between the query and the plurality of relevant documents, respectively. ### Claim 12 (depends on 11) The medium of claim 11, wherein the one or more parameters characterize a curve formed by connecting the plurality of similarities. ### Claim 13 (depends on 8) The medium of claim 8, wherein the information, when read by the machine, further causing the machine to compute at least one of: a novelty score characterizing novelty of the query with respect to the plurality of relevant documents; and a non-obviousness score characterizing non-obviousness of the query with respect to the plurality of relevant documents. ### Claim 14 (depends on 6) The medium of claim 6, wherein the information, when read by the machine, further causing the machine to perform the following: computing a distinction index of the query based on the novelty score and/or the non-obviousness score. ### Claim 15 A system for assessing semantic distance, comprising: a document representation constructor implemented by a processor and configured for obtaining a first query representation of an input query capturing characteristics of the query; a query representation constructor implemented by the processor and configured for obtaining a second query representation of the input query capturing the characteristics of the query; a dynamic norm determiner implemented by the processor and configured for determining a dynamic norm based on a discrepancy between the first and the second query representations, wherein the dynamic norm is indicative of a bias between the document and query representation constructors; a relevant document identifier implemented by the processor and configured for identifying, based on the second query representation, a plurality of documents relevant to the input query, and for each of the plurality of documents, computing a similarity between the document represented by a document representation obtained via the document representation constructor and the query represented by the second query representation obtained via the query representation constructor; and a semantic distance determiner implemented by the processor and configured for determining a semantic distance between the input query and each of the plurality of documents based on the dynamic norm, wherein the semantic distance is derived by adjusting a distance between the query and the document based on the dynamic norm to address the bias. ### Claim 16 (depends on 15) The system of claim 15, wherein the dynamic norm is determined by: computing a similarity between the first and the second query representations; and determining the dynamic norm based on the similarity. ### Claim 17 (depends on 16) The system of claim 16, wherein the semantic distance is determined by: computing a similarity between the query and each of the plurality of relevant documents; determining a statistic based on the similarities between the query and the plurality of relevant documents; and obtaining the semantic distance based on the statistic and the dynamic norm. ### Claim 18 (depends on 15) The system of claim 15, wherein the semantic distance is determined further based on a confidence score, which is estimated based on one or more parameters characterizing a distribution pattern of a plurality of similarities measured between the query and the plurality of relevant documents, respectively. ### Claim 19 (depends on 18) The system of claim 18, wherein the one or more parameters characterize a curve formed by connecting the plurality of similarities. ### Claim 20 (depends on 15) The system of claim 15, further comprising: a novelty assessment unit implemented by the processor and configured for determining a novelty score characterizing novelty of the input query with respect to the plurality of documents; and a non-obviousness assessment unit implemented by the processor and configured for determining a non-obviousness score characterizing non-obviousness of the input query with respect to the plurality of documents, wherein the novelty score and the non-obviousness score are determined based on semantic distances between the input query and the respective document of the plurality of documents. ### Claim 21 (depends on 20) The system of claim 20, further comprising: a distinction index determiner implemented by the processor and configured for computing a distinction index of the input query based on the novelty score and/or the non-obviousness score.