# US 2023/0054726 A1 · Query-focused extractive text summarization of textual data > Agents: this data is available as JSON from the patentgrep API and MCP server — see https://patentgrep.com/llms.txt - Number: US 2023/0054726 A1 - Document ID: US20230054726A1 - Status: Published application - Kind: A1 - Application number: 17/405,555 - Filed: 2021-08-18 - Published: 2023-02-23 - Earliest claimed: 2021-08-18 - Assignees: Optum, Inc. - Applicants: Optum, Inc. - Inventors: Suman Roy; Vijay Varma Malladi; Gaurav Ranjan - Web page: https://patentgrep.com/patent/US20230054726A1 ## CPC - G06F 40/284: Lexical analysis, e.g. tokenisation or collocates - G06F 16/3329: Natural language query formulation - G06F 16/3334: {Selection or weighting of terms from queries, including natural language queries} - G06F 16/345: {Summarisation for human users} - G06F 16/35: Clustering; Classification - G06F 16/383: using metadata automatically derived from the content - G06F 40/30: Semantic analysis - G06N 3/044: Recurrent networks, e.g. Hopfield networks - G06N 3/0442: characterised by memory or gating, e.g. long short-term memory [LSTM] or gated recurrent units [GRU] - G06N 3/08: Learning methods - G06N 3/088: Non-supervised learning, e.g. competitive learning - G10L 15/26: Speech to text systems (G10L15/08 takes precedence) - H04L 51/02: using automatic reactions or user delegation, e.g. automatic replies or chatbot-generated messages - H04L 12/1831: {Tracking arrangements for later retrieval, e.g. recording contents, participants activities or behavior, network status} - G06F: ELECTRIC DIGITAL DATA PROCESSING (computer systems based on specific computational models G06N) - G06N: COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS - G10L: SPEECH ANALYSIS TECHNIQUES OR SPEECH SYNTHESIS; SPEECH RECOGNITION; SPEECH OR VOICE PROCESSING TECHNIQUES; SPEECH OR AUDIO CODING OR DECODING - H04L: TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION (arrangements common to telegraphic and telephonic communication H04M) ## Abstract Various embodiments provide methods, apparatus, systems, computing entities, and/or the like, for providing a summarization of a conversation, such as a telephonic conversation. In an embodiment, a method is provided. The method comprises receiving an input data object comprising textual data of a conversation, the textual data comprising sentence-level tokens. The method further comprises classifying some sentence-level tokens as interrogative sentence-level tokens, and identifying subtopic portions of the textual data, each interrogative sentence-level token located within one subtopic portion. The method further comprises determining whether an interrogative sentence-level token is substantially similar to one of a plurality of target queries, and for such interrogative sentence-level tokens, selecting sentence-level tokens from a subtopic portion corresponding to the such interrogative sentence-level tokens. The method then comprises generating a summarization data object comprising the selected sentence-level tokens for each interrogative sentence-level token substantially similar to a target query and performing summarization-based actions. ## Claims ### Claim 1 A computer-implemented method for providing a summarization of a conversation, the computer-implemented method comprising: receiving, by a processor, an input data object comprising textual data of a conversation, wherein the textual data comprises a plurality of sentence-level tokens and each sentence-level token comprises one or more word-level tokens; generating and assigning, by the processor, an interrogative classification for one or more sentence-level tokens of the plurality of sentence-level tokens, wherein the interrogative classification is determined based at least in part on evaluating the one or more word-level tokens of a sentence-level token; identifying, by the processor, one or more subtopic portions of the textual data, wherein each subtopic portion comprises a portion of the plurality of sentence-level tokens and at least one interrogative sentence-level token, and wherein each interrogative sentence-level token is located within one subtopic portion; for each interrogative sentence-level token, determining, by the processor, whether the interrogative sentence-level token is substantially similar to a target query of a plurality of target queries; for each particular interrogative sentence-level token determined to be substantially similar to a target query, selecting, by the processor, one or more sentence-level tokens in a subtopic portion corresponding to the particular interrogative sentence-level token; generating, by the processor, a summarization data object comprising the selected one or more sentence-level tokens for each particular interrogative sentence-level token determined to be substantially similar to a target query; and performing, by the processor, one or more summarization-based actions based at least in part on the summarization data object. ### Claim 2 (depends on 1) The computer-implemented method of claim 1, wherein selecting one or more sentence-level tokens in a subtopic portion corresponding to the particular interrogative sentence-level token comprises: generating a first characterization score for each sentence-level token occurring after the particular interrogative sentence-level token in the subtopic portion, wherein the first characterization score of a sentence-level token is indicative of an informational value of the sentence-level token; generating a second characterization score for each sentence-level token occurring after the particular interrogative sentence-level token in the subtopic portion, wherein the second characterization score of a sentence-level token is indicative of a linguistic quality of the sentence-level token; generating an aggregate characterization score for each sentence-level token occurring after the particular interrogative sentence-level token in the subtopic portion, wherein the aggregate characterization score for a sentence-level token is based at least in part on the first characterization score and the second characterization score of the sentence-level token; and selecting one or more sentence-level tokens occurring after the particular interrogative sentence-level token in the subtopic portion based at least in part on the aggregate characterization scores. ### Claim 3 (depends on 1) The computer-implemented method of claim 1, wherein the summarization data object comprises the plurality of target queries, wherein a target query of the plurality of target queries is provided with one or more selected sentence-level tokens for each particular interrogative sentence-level token determined to be substantially similar to the target query. ### Claim 4 (depends on 1) The computer-implemented method of claim 1, wherein each target query of the plurality of target queries is associated with one or more equivalent variants, and determining whether the interrogative sentence-level token is substantially similar to a target query comprises determining whether the interrogative sentence-level token is substantially similar to an equivalent variant for each of the one or more equivalent variants. ### Claim 5 (depends on 1) The computer-implemented method of claim 1, wherein identifying one or more subtopic portions of the textual data comprises generating one or more paragraph-level tokens in the textual data, each paragraph-level token comprising at least one sentence-level token and beginning with an interrogative sentence-level token. ### Claim 6 (depends on 5) The computer-implemented method of claim 5, wherein a subtopic portion comprises one or more paragraph-level tokens. ### Claim 7 (depends on 1) The computer-implemented method of claim 1, wherein an interrogative classification is generated and assigned to a sentence-level token based at least in part on providing the one or more word-level tokens of the sentence-level token to a self-attention based bidirectional long short term memory (LSTM) model. ### Claim 8 (depends on 1) The computer-implemented method of claim 1, wherein receiving the input data object comprising textual data comprises performing pre-processing operations on the textual data to remove one or more word-level tokens describing stop words based at least in part on a configurable list of stop words. ### Claim 9 (depends on 1) The computer-implemented method of claim 1, wherein the textual data of the input data object is transcribed textual data from a telephonic conversation. ### Claim 10 An apparatus comprising at least one processor and at least one memory comprising computer program code, the at least one memory and the computer program code configured to, with the at least one processor, cause the apparatus to: receive an input data object comprising textual data of a conversation, wherein the textual data comprises a plurality of sentence-level tokens and each sentence-level token comprises one or more word-level tokens; generate and assign an interrogative classification for one or more sentence-level tokens of the plurality of sentence-level tokens, wherein the interrogative classification is determined based at least in part on evaluating the one or more word-level tokens of a sentence-level token; identify one or more subtopic portions of the textual data, wherein each subtopic portion comprises a portion of the plurality of sentence-level tokens and at least one interrogative sentence-level token, and wherein each interrogative sentence-level token is located within one subtopic portion; for each interrogative sentence-level token, determine whether the interrogative sentence-level token is substantially similar to a target query of a plurality of target queries; for each particular interrogative sentence-level token determined to be substantially similar to a target query, select one or more sentence-level tokens in a subtopic portion corresponding to the particular interrogative sentence-level token; generate a summarization data object comprising the selected one or more sentence-level tokens for each particular interrogative sentence-level token determined to be substantially similar to a target query; and perform one or more summarization-based actions based at least in part on the summarization data object. ### Claim 11 (depends on 10) The apparatus of claim 10, wherein selecting one or more sentence-level tokens in a subtopic portion corresponding to the particular interrogative sentence-level token comprises: generating a first characterization score for each sentence-level token occurring after the particular interrogative sentence-level token in the subtopic portion, wherein the first characterization score of a sentence-level token is indicative of an informational value of the sentence-level token; generating a second characterization score for each sentence-level token occurring after the particular interrogative sentence-level token in the subtopic portion, wherein the second characterization score of a sentence-level token is indicative of a linguistic quality of the sentence-level token; generating an aggregate characterization score for each sentence-level token occurring after the particular interrogative sentence-level token in the subtopic portion, wherein the aggregate characterization score for a sentence-level token is based at least in part on the first characterization score and the second characterization score of the sentence-level token; and selecting one or more sentence-level tokens occurring after the particular interrogative sentence-level token in the subtopic portion based at least in part on the aggregate characterization scores. ### Claim 12 (depends on 10) The apparatus of claim 10, wherein the summarization data object comprises the plurality of target queries, wherein a target query of the plurality of target queries is provided with one or more selected sentence-level tokens for each particular interrogative sentence-level token determined to be substantially similar to the target query. ### Claim 13 (depends on 10) The apparatus of claim 10, wherein each target query of the plurality of target queries is associated with one or more equivalent variants, and determining whether the interrogative sentence-level token is substantially similar to a target query comprises determining whether the interrogative sentence-level token is substantially similar to an equivalent variant for each of the one or more equivalent variants. ### Claim 14 (depends on 10) The apparatus of claim 10, wherein identifying one or more subtopic portions of the textual data comprises generating one or more paragraph-level tokens in the textual data, each paragraph-level token comprising at least one sentence-level token and beginning with an interrogative sentence-level token. ### Claim 15 (depends on 14) The apparatus of claim 14, wherein a subtopic portion comprises one or more paragraph-level tokens. ### Claim 16 (depends on 10) The apparatus of claim 10, wherein an interrogative classification is generated and assigned to a sentence-level token based at least in part on providing the one or more word-level tokens of the sentence-level token to a self-attention based bidirectional long short term memory (LSTM) model. ### Claim 17 (depends on 10) The computer-implemented method of claim 10, wherein receiving the input data object comprising textual data comprises performing pre-processing operations on the textual data to remove one or more word-level tokens describing stop words based at least in part on a configurable list of stop words. ### Claim 18 (depends on 10) The computer-implemented method of claim 10, wherein the textual data of the input data object is transcribed textual data from a telephonic conversation. ### Claim 19 A computer program product comprising at least one computer-readable storage medium having computer-readable program code portions stored therein, the computer-readable program code portions comprising executable portions configured to cause at least one processor to: receive an input data object comprising textual data of a conversation, wherein the textual data comprises a plurality of sentence-level tokens and each sentence-level token comprises one or more word-level tokens; generate and assign an interrogative classification for one or more sentence-level tokens of the plurality of sentence-level tokens, wherein the interrogative classification is determined based at least in part on evaluating the one or more word-level tokens of a sentence-level token; identify one or more subtopic portions of the textual data, wherein each subtopic portion comprises a portion of the plurality of sentence-level tokens and at least one interrogative sentence-level token, and wherein each interrogative sentence-level token is located within one subtopic portion; for each interrogative sentence-level token, determine whether the interrogative sentence-level token is substantially similar to a target query of a plurality of target queries; for each particular interrogative sentence-level token determined to be substantially similar to a target query, select one or more sentence-level tokens in a subtopic portion corresponding to the particular interrogative sentence-level token; generate a summarization data object comprising the selected one or more sentence-level tokens for each particular interrogative sentence-level token determined to be substantially similar to a target query; and perform one or more summarization-based actions based at least in part on the summarization data object. ### Claim 20 (depends on 19) The computer program product of claim 19, wherein an interrogative classification is generated and assigned to a sentence-level token based at least in part on providing the one or more word-level tokens of the sentence-level token to a self-attention based bidirectional long short term memory (LSTM) model.