# US 11,308,287 B1 · Background conversation analysis for providing a real-time feedback > Agents: this data is available as JSON from the patentgrep API and MCP server — see https://patentgrep.com/llms.txt - Number: US 11,308,287 B1 - Document ID: US11308287B1 - Status: Granted - Kind: B1 - Application number: 16/948,807 - Filed: 2020-10-01 - Published: 2022-04-07 - Granted: 2022-04-19 - Earliest claimed: 2020-10-01 - Assignees: International Business Machines Corporation - Applicants: International Business Machines Corporation - Inventors: Nadiya Kochura; Jonathan D. Dunne; Fang Lu - Web page: https://patentgrep.com/patent/US11308287B1 ## CPC - G06F 40/30: Semantic analysis - G06F 40/56: Natural language generation - G06F 40/289: Phrasal analysis, e.g. finite state techniques or chunking - G06F 40/35: Discourse or dialogue representation - G06F: ELECTRIC DIGITAL DATA PROCESSING (computer systems based on specific computational models G06N) ## Abstract A computing device that receives a real-time chat discourse. The computing device analyses the real-time chat discourse by consecutively applying a topic analysis technique, a corpus linguistics technique and a cosine similarity technique. The computing device derives a discourse decision forking component (DDFC) based on comparing the analyzed real-time chat discourse to a similarity threshold value and determines one or more discourse forks using the DDFC. ## Claims ### Claim 1 A processor-implemented method for a chat discourse decision forking, the method comprising: receiving a real-time chat discourse; analyzing the real-time chat discourse, wherein the real-time chat discourse is analyzed by consecutively applying a topic analysis technique to determine one or more topics in the real-time chat discourse, a corpus linguistics technique and a cosine similarity technique; deriving a discourse decision forking component (DDFC) based on comparing the analyzed real-time chat discourse to a similarity threshold value; and determining one or more discourse forks using the DDFC. ### Claim 2 (depends on 1) The method of claim 1, further comprising: displaying the one or more determined discourse forks using a graphic user interface component as a topic tree wherein the topic tree comprises topics as branches of the topic tree, wherein the topics derived from the topic analysis technique and arranged based on the determined one or more discourse forks. ### Claim 3 (depends on 1) The method of claim 1, further comprising: generating a message in the chat discourse based on the determined one or more discourse forks. ### Claim 4 (depends on 1) The method of claim 1, wherein the topic analysis technique is based on a Latent Dirichlet Allocation (LDA) approach. ### Claim 5 (depends on 1) The method of claim 1, wherein the corpus linguistic technique transfers the one or more topics into a set of multi-dimensional vectors. ### Claim 6 (depends on 5) The method of claim 5, wherein deriving the DDFC based on comparing the analyzed real-time chat discourse to the similarity threshold value comprises: determining a similarity value between the set of multi-dimensional vectors; and based on comparing the similarity value to the similarity threshold value, deriving the DDFC. ### Claim 7 (depends on 1) The method of claim 1, wherein the similarity threshold value is determined based on a domain, wherein the domain derived from the chat discourse. ### Claim 8 A computer system for a chat discourse decision forking, the computer system comprising: one or more processors, one or more computer-readable memories, one or more computer-readable tangible storage medium, and program instructions stored on at least one of the one or more tangible storage medium for execution by at least one of the one or more processors via at least one of the one or more memories, wherein the computer system is capable of performing a method comprising: receiving a real-time chat discourse; analyzing the real-time chat discourse, wherein the real-time chat discourse is analyzed by consecutively applying a topic analysis technique to determine one or more topics in the real-time chat discourse, a corpus linguistics technique and a cosine similarity technique; deriving a discourse decision forking component (DDFC) based on comparing the analyzed real-time chat discourse to a similarity threshold value; and determining one or more discourse forks using the DDFC. ### Claim 9 (depends on 8) The computer system of claim 8, further comprising displaying the one or more determined discourse forks using a graphic user interface component as a topic tree wherein the topic tree comprises topics as branches of the topic tree, wherein the topics derived from the topic analysis technique and arranged based on the determined one or more discourse forks. ### Claim 10 (depends on 8) The computer system of claim 8, further comprising generating a message in the chat discourse based on the determined one or more discourse forks. ### Claim 11 (depends on 8) The computer system of claim 8, wherein the topic analysis technique is based on a Latent Dirichlet Allocation (LDA) approach. ### Claim 12 (depends on 8) The computer system of claim 8, wherein the corpus linguistic technique transfers the one or more topics into a set of multi-dimensional vectors. ### Claim 13 (depends on 12) The computer system of claim 12, wherein deriving the DDFC based on comparing the analyzed real-time chat discourse to the similarity threshold value comprises: determining a similarity value between the set of multi-dimensional vectors; and based on comparing the similarity value to the similarity threshold value deriving the DDFC. ### Claim 14 (depends on 8) The computer system of claim 8, wherein the similarity threshold value is determined based on a domain, wherein the domain derived from the chat discourse. ### Claim 15 A computer program product for a chat discourse decision forking, the computer program product comprising: one or more computer-readable tangible storage medium and program instructions stored on at least one of the one or more tangible storage medium, the program instructions executable by a processor, the program instructions comprising: program instructions to receive a real-time chat discourse; program instructions to analyze the real-time chat discourse, wherein the real-time chat discourse is analyzed by consecutively applying a topic analysis technique to determine one or more topics in the real-time chat discourse, a corpus linguistics technique and a cosine similarity technique; program instructions to derive a discourse decision forking component (DDFC) based on comparing the analyzed real-time chat discourse to a similarity threshold value; and program instructions to determine one or more discourse forks using the DDFC. ### Claim 16 (depends on 15) The computer program product of claim 15, further comprising program instructions to display the one or more determined discourse forks using a graphic user interface component as a topic tree wherein the topic tree comprises topics as branches of the topic tree, wherein the topics derived from the topic analysis technique and arranged based on the determined one or more discourse forks. ### Claim 17 (depends on 15) The computer program product of claim 15, further comprising program instructions to generate a message in the chat discourse based on the determined one or more discourse forks. ### Claim 18 (depends on 15) The computer program product of claim 15, wherein the topic analysis technique is based on a Latent Dirichlet Allocation (LDA) approach. ### Claim 19 (depends on 15) The computer program product of claim 15, wherein the corpus linguistic technique transfers the one or more topics into a set of multi-dimensional vectors. ### Claim 20 (depends on 19) The computer program product of claim 19, wherein program instructions to derive the DDFC based on program instructions to compare the analyzed real-time chat discourse to the similarity threshold value comprises: program instructions to determine a similarity value between the set of multi-dimensional vectors; and based on comparing the similarity value to the similarity threshold value program instructions to derive the DDFC.