# US 10,120,865 B1 · Techniques for automatic generation of natural language text > Agents: this data is available as JSON from the patentgrep API and MCP server — see https://patentgrep.com/llms.txt - Number: US 10,120,865 B1 - Document ID: US10120865B1 - Status: Granted - Kind: B1 - Application number: 15/201,930 - Filed: 2016-07-05 - Granted: 2018-11-06 - Earliest claimed: 2013-07-17 - Assignees: Yseop SA - Applicants: Yseop SA - Inventors: Alain Kaeser; Emmanuel Vignon; Ludan Stoeckle - Web page: https://patentgrep.com/patent/US10120865B1 ## CPC - G06F 40/35: Discourse or dialogue representation - G06F 40/253: Grammatical analysis; Style critique - G06F 40/40: Processing or translation of natural language (natural language analysis G06F40/20; semantic analysis G06F40/30) - G06F 40/56: Natural language generation - G06F 40/205: Parsing - G06F 40/289: Phrasal analysis, e.g. finite state techniques or chunking - G06F: ELECTRIC DIGITAL DATA PROCESSING (computer systems based on specific computational models G06N) ## Abstract Techniques for use in connection with a system for automatically generating text. Techniques include accessing information specifying at least one referential expression for at least a first referent and at least one anaphoric expression for at least the first referent; accessing a template that includes human-language text and a first tag that serves as a placeholder for a first text portion including a reference to at least the first referent; automatically identifying, using at least one system rule and at least one processor, text to use for the first text portion at least in part by determining whether to use as the text for the first text portion the at least one referential expression or the at least one anaphoric expression; and automatically generating output text including the human-language text and the identified text for the first text portion. ## Claims ### Claim 1 A system, comprising: at least one hardware processor; and at least one non-transitory computer-readable storage medium storing instructions that, when executed by the at least one hardware processor, cause the at least one hardware processor to perform: accessing information specifying at least one referential expression for a first referent and at least one anaphoric expression for the first referent; accessing a template including human-language text, a first tag that serves as a placeholder for a first text portion including a first reference to the first referent, a second tag that precedes the first tag in the template and serves as a placeholder for a second text portion including a second reference to a second referent different from the first referent, and a third tag that precedes the second tag in the template and serves as a placeholder for a third text portion including a third reference to the first referent; determining whether using the at least one anaphoric expression for the first text portion would result in the insertion of an ambiguous reference in the output text by determining whether the first and second referents are of the same type, whether the first and second referents are mutually ambiguous, or whether the first and second referents are of the same type and whether the first and second referents are mutually ambiguous; identifying text to use for the first text portion based on results of determining whether using the at least one anaphoric expression would result in the insertion of an ambiguous reference in the output text, the identifying comprising: when it is determined that using the at least one anaphoric expression as the text for the first text portion would result in the insertion of an ambiguous reference, making a determination to use the at least one referential expression as the text for the first text portion; and when it is determined that using the at least one anaphoric expression as the text for the first text portion would not result in the insertion of an ambiguous reference, making a determination to use the at least one anaphoric expression as the text for the first text portion; automatically generating output text including the human-language text and the identified text for the first text portion; and presenting, via a device, the automatically generated output text to a user. ### Claim 2 (depends on 1) The system of claim 1, wherein the at least one hardware processor is configured to determine whether using the at least one anaphoric expression for the first text portion would result in the insertion of an ambiguous reference in the output text further based on at least one of a grammatical number of the at least one anaphoric expression, a gender of the at least one anaphoric expression, whether the at least one anaphoric expression is a pronoun, and language of the human-language text in the template. ### Claim 3 (depends on 1) The system of claim 1, wherein the first reference in the first text portion to the first referent matches the second reference to the second referent, and wherein the at least one anaphoric expression comprises a plural anaphoric expression for referring simultaneously to both the first referent and the second referent. ### Claim 4 (depends on 1) The system of claim 1, wherein the at least one anaphoric expression includes an anaphoric expression that is not a pronoun. ### Claim 5 (depends on 1) The system of claim 1, wherein the instructions cause the at least one hardware processor to perform determining whether using the at least one anaphoric expression for the first text portion would result in the insertion of an ambiguous reference in the output text by determining whether the first and second referents are of the same type. ### Claim 6 (depends on 1) The system of claim 1, wherein the instructions further cause the at least one hardware processor to access at least one user-specified rule for the first tag, wherein the at least one hardware processor identifies text based at least in part on the at least one user-specified rule. ### Claim 7 (depends on 1) The system of claim 1, wherein the first referent comprises a plurality of referents each having a gender among a plurality of genders, and wherein the information specifies different anaphoric expressions for different genders in the plurality of genders. ### Claim 8 A system, comprising: at least one hardware processor; and at least one non-transitory computer-readable storage medium storing instructions that, when executed by the at least one hardware processor, cause the at least one hardware processor to perform: accessing information specifying at least one referential expression for a first referent and at least one anaphoric expression for the first referent; accessing a template including human-language text, a first tag that serves as a placeholder for a first text portion including a first reference to the first referent, and a second tag that serves as a placeholder for a second text portion including a second reference to a second referent; automatically identifying, using software encoding at least one system rule, text to use for the first text portion at least in part by determining whether to use, for the first text portion, the at least one referential expression or the at least one anaphoric expression, wherein the determining comprises: determining to use the at least one anaphoric expression as the text for the first text portion when the system has determined to use the at least one referential expression to replace the second tag in the template, and when the second tag is within a threshold distance of the first tag in the template; automatically generating output text including the human-language text and the identified text for the first text portion; and presenting, via a device, the automatically generated output text to a user. ### Claim 9 (depends on 8) The system of claim 8, wherein the at least one hardware processor is further configured to determine whether using the at least one anaphoric expression for the first text portion would result in the insertion of an ambiguous reference in the output text based on at least one of a grammatical number of the at least one anaphoric expression, a gender of the at least one anaphoric expression, whether the at least one anaphoric expression is a pronoun, and language of the human-language text in the template. ### Claim 10 (depends on 8) The system of claim 8, wherein the first reference in the first text portion to the first referent matches the second reference to the second referent, and wherein the at least one anaphoric expression comprises a plural anaphoric expression for referring simultaneously to both the first referent and the second referent. ### Claim 11 (depends on 8) The system of claim 8, wherein the at least one anaphoric expression includes an anaphoric expression that is not a pronoun. ### Claim 12 (depends on 11) The system of claim 11, wherein the instructions further cause the at least one hardware processor to perform determining whether using the at least one anaphoric expression for the first text portion would result in the insertion of an ambiguous reference in the output text by determining whether the first and second referents are of the same type. ### Claim 13 (depends on 8) The system of claim 8, wherein the instructions further cause the at least one hardware processor to access at least one user-specified rule for the first tag, wherein the at least one hardware processor automatically identifies text based at least in part on the at least one user-specified rule. ### Claim 14 (depends on 8) The system of claim 8, wherein the first referent comprises a plurality of referents each having a gender among a plurality of genders, and wherein the information specifies different anaphoric expressions for different genders in the plurality of genders. ### Claim 15 A system, comprising: at least one hardware processor; and at least one non-transitory computer-readable storage medium storing instructions that, when executed by the at least one hardware processor, cause the at least one hardware processor to perform: accessing a template including human-language text and a first tag that serves as a placeholder for a first text portion including a first reference to a first referent, wherein the first reference is also a reference to a second referent different from the first referent; accessing information specifying a first plural referential expression for referring simultaneously to both the first referent and the second referent and a first plural anaphoric expression for referring simultaneously to both the first referent and the second referent; automatically identifying text to use for the first text portion at least in part by determining whether to use, for the first text portion, the plural referential expression or the plural anaphoric expression; automatically generating output text including the human-language text and the identified text for the first text portion; and presenting, via a device, the automatically generated output text to a user. ### Claim 16 (depends on 15) The system of claim 15, wherein the at least one hardware processor is further configured to determine whether using the first plural anaphoric expression for the first text portion would result in the insertion of an ambiguous reference in the output text based on whether the first plural anaphoric expression is a pronoun. ### Claim 17 (depends on 15) The system of claim 15, wherein the instructions further cause the at least one hardware processor to perform determining whether using the at least one anaphoric expression for the first text portion would result in the insertion of an ambiguous reference in the output text by determining whether the first and second referents are of the same type. ### Claim 18 (depends on 15) The system of claim 15, wherein the instructions further cause the at least one hardware processor to perform determining whether using the at least one anaphoric expression for the first text portion would result in the insertion of an ambiguous reference in the output text by determining whether the first and second referents are mutually ambiguous. ### Claim 19 (depends on 15) The system of claim 15, wherein the instructions further cause the at least one hardware processor to access at least one user-specified rule for the first tag, wherein the at least one hardware processor automatically identifies text based at least in part on the at least one user-specified rule. ### Claim 20 (depends on 15) The system of claim 15, wherein the first plural anaphoric expression is not a pronoun.