# US 10,769,373 B2 · Contextual validation of synonyms in otology driven natural language processing > Agents: this data is available as JSON from the patentgrep API and MCP server — see https://patentgrep.com/llms.txt - Number: US 10,769,373 B2 - Document ID: US10769373B2 - Status: Granted - Kind: B2 - Application number: 16/553,796 - Filed: 2019-08-28 - Published: 2019-12-19 - Granted: 2020-09-08 - Earliest claimed: 2013-11-20 - Assignees: International Business Machines Corporation - Applicants: International Business Machines Corporation - Inventors: Stephen J. Edwards; Ahmed M. Nassar; Craig M. Trim; Albert T. Wong - Web page: https://patentgrep.com/patent/US10769373B2 ## CPC - G06F 40/247: Thesauruses; Synonyms - G06F 40/169: Annotation, e.g. comment data or footnotes - G06F 40/205: Parsing - G06F 40/284: Lexical analysis, e.g. tokenisation or collocates - G06F 40/30: Semantic analysis - G06F: ELECTRIC DIGITAL DATA PROCESSING (computer systems based on specific computational models G06N) ## Abstract Embodiments described herein provide approaches for validating synonyms in ontology driven natural language processing. Specifically, an approach is provided for receiving a user input containing a token, structuring the user input into a semantic model comprising a set of classes each containing a set of related permutations of the token, designating the token as a synonym of one of the set of related permutations, annotating the token with a class from the set of classes corresponding to the one of the set of related permutations, and validating the annotation of the token by determining an accuracy of the designation of the token as a synonym of the one of the set of related permutations. In one embodiment, the accuracy is determined by quantifying a linear distance between the token and a contextual token also within the user input, and comparing the linear distance to a pre-specified linear distance limit. ## Claims ### Claim 1 A method for providing contextual validation of synonyms in ontology driven natural language processing, the method comprising the computer-implemented steps of: determining, via at least one computing device, based on a linear sequence of symbols in electronic text, a token that identifies a linguistic unit of the electronic text, the linguistic unit comprising at least one of a word, a punctuation symbol, a number, or a letter; structuring, via at least one computing device, the user input into a semantic model comprising a set of classes each containing a set of related permutations of the token, wherein the semantic model is stored as data in memory of the at least one computing device; quantifying, via at least one computing device, a linear distance between the token and a contextual token within the user input, wherein the linear distance is a quantity of additional tokens in between the token and the contextual token; comparing, via at least one computing device, the linear distance to a pre-specified linear distance limit to a number of words; when, based on the comparing, the quantified linear distance is within the pre-specified linear distance limit to the number of words, assigning a high confidence level to an annotation of the token, and validating the annotation based on the high confidence level, wherein the validating the annotation comprises restructuring, via at least one computing device, the semantic model to include a knowledge structure containing the contextual token, the linear distance, the pre-specified linear distance limit, and the designation of the token as a synonym of the one of the set of related permutations; and when, based on the comparing, the quantified linear distance is not within the pre-specified linear distance limit to the number of words, assigning a low confidence level to the annotation. ### Claim 2 (depends on 1) The method according to claim 1, wherein the token is defined by a string or word delimited on both sides by spaces or punctuation. ### Claim 3 (depends on 1) The method according to claim 1, wherein the token is defined as an item that is linguistically significant. ### Claim 4 (depends on 1) The method according to claim 1, wherein the pre-specified linear distance limit is user-defined. ### Claim 5 (depends on 2) The method according to claim 2, wherein the linear distance corresponds to a quantity of tokens. ### Claim 6 (depends on 1) The method according to claim 1, further comprising the computer-implemented step of parsing the user input. ### Claim 7 (depends on 1) The method according to claim 1, wherein a solution service provider deploys a computer infrastructure operable to perform the method. ### Claim 8 A system for providing contextual validation of synonyms in ontology driven natural language processing (NLP), the system comprising: a memory medium comprising instructions; a bus coupled to the memory medium; and a processor coupled to a NLP engine via the bus that when executing the instructions causes the system to: determine, based on a linear sequence of symbols in electronic text, a token that identifies a linguistic unit of the electronic text, the linguistic unit comprising at least one of a word, a punctuation symbol, a number, or a letter; structure the user input into a semantic model comprising a set of classes each containing a set of related permutations of the token, wherein the semantic model is stored as data in memory of the at least one computing device; quantify a linear distance between the token and a contextual token within the user input, wherein the linear distance is a quantity of additional tokens in between the token and the contextual token; compare the linear distance to a pre-specified linear distance limit to a number of words; when the quantified linear distance is within the pre-specified linear distance limit to the number of words, assign a high confidence level to an annotation of the token, and validate the annotation based on the high confidence level, wherein the validating the annotation comprises restructuring, via at least one computing device, the semantic model to include a knowledge structure containing the contextual token, the linear distance, the pre-specified linear distance limit, and the designation of the token as a synonym of the one of the set of related permutations; and when, based on the comparing, the quantified linear distance is not within the pre-specified linear distance limit to the number of words, assign a low confidence level to the annotation. ### Claim 9 (depends on 8) The system according to claim 8, wherein the token is defined by a string or word delimited on both sides by spaces or punctuation. ### Claim 10 (depends on 8) The system according to claim 8, wherein the token is defined as an item that is linguistically significant. ### Claim 11 (depends on 8) The system according to claim 8, wherein the pre-specified linear distance limit is user-defined. ### Claim 12 (depends on 8) The system according to claim 8, the linear distance corresponding to a quantity of tokens. ### Claim 13 (depends on 8) The system according to claim 8, further comprising instructions to parse the user input. ### Claim 14 A computer-readable storage device storing computer instructions, which when executed, enables a computer system to provide contextual validation of synonyms in ontology driven natural language processing (NLP), the computer instructions comprising: determine, based on a linear sequence of symbols in electronic text, a token that identifies a linguistic unit of the electronic text, the linguistic unit comprising at least one of a word, a punctuation symbol, a number, or a letter; structure the user input into a semantic model comprising a set of classes each containing a set of related permutations of the token, wherein the semantic model is stored as data in memory of the at least one computing device; quantify a linear distance between the token and a contextual token within the user input, wherein the linear distance is a quantity of additional tokens in between the token and the contextual token; compare the linear distance to a pre-specified linear distance limit to a number of words; when the quantified linear distance is within the pre-specified linear distance limit to the number of words, assign a high confidence level to an annotation of the token, and validate the annotation based on the high confidence level, wherein the validating the annotation comprises restructuring, via at least one computing device, the semantic model to include a knowledge structure containing the contextual token, the linear distance, the pre-specified linear distance limit, and the designation of the token as a synonym of the one of the set of related permutations; and when, based on the comparing, the quantified linear distance is not within the pre-specified linear distance limit to the number of words, assign a low confidence level to the annotation. ### Claim 15 (depends on 14) The computer-readable storage device according to claim 14, wherein the token is defined by a string or word delimited on both sides by spaces or punctuation. ### Claim 16 (depends on 14) The computer-readable storage device according to claim 14, wherein the token is defined as an item that is linguistically significant. ### Claim 17 (depends on 14) The computer-readable storage device according to claim 14, wherein the pre-specified linear distance limit is user-defined. ### Claim 18 (depends on 14) The computer-readable storage device according to claim 14, the linear distance corresponding to a quantity of tokens. ### Claim 19 (depends on 14) The computer-readable storage device according to claim 14, the computer instructions further comprising parsing the user input. ### Claim 20 The computer-readable storage device according to claim, 19, wherein the parsing is performed prior to the structuring.