# US 2023/0057760 A1 · Constructing conclusive answers for autonomous agents > Agents: this data is available as JSON from the patentgrep API and MCP server — see https://patentgrep.com/llms.txt - Number: US 2023/0057760 A1 - Document ID: US20230057760A1 - Status: Published application - Kind: A1 - Application number: 17/974,335 - Filed: 2022-10-26 - Published: 2023-02-23 - Earliest claimed: 2018-10-16 - Assignees: Oracle International Corporation - Applicants: Oracle International Corporation - Inventors: Boris Galitsky - Web page: https://patentgrep.com/patent/US20230057760A1 ## CPC - G06F 40/211: Syntactic parsing, e.g. based on context-free grammar [CFG] or unification grammars - G06F 16/2246: {Trees, e.g. B+trees} - G06F 16/3329: Natural language query formulation - G06F 16/9027: {Trees} - G06F 16/953: Querying, e.g. by the use of web search engines - G06F 40/279: Recognition of textual entities - G06F 40/56: Natural language generation - G06N 5/01: Dynamic search techniques; Heuristics; Dynamic trees; Branch-and-bound - G06N 5/04: Inference or reasoning models - G06N 20/00: Machine learning - G06N 7/01: Probabilistic graphical models, e.g. probabilistic networks - G06F: ELECTRIC DIGITAL DATA PROCESSING (computer systems based on specific computational models G06N) - G06N: COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS ## Abstract Techniques are described herein for enabling autonomous agents to generate conclusive answers. An example of a conclusive answer is text that addresses concerns of a user who is interacting with an autonomous agent. For example, an autonomous agent interacts with a user device, answering user utterances, for example questions or concerns. Based on the interactions, the autonomous agent determines that a conclusive answer is appropriate. The autonomous agent formulates the conclusive answer, which addresses multiple user utterances. The conclusive answer provided to the user device. ## Claims ### Claim 1 A method of computationally fortifying an answer using syntactic parse trees, the method comprising: accessing a seed sentence comprising a first plurality of text fragments; obtaining a search result by providing, to a search engine, at least one of the first plurality of text fragments, wherein the search result comprises a second plurality of text fragments; generating, from the search result, a communicative discourse tree, wherein generating the communicative discourse tree comprises: creating a discourse tree from the second plurality of text fragments, wherein the discourse tree comprises a plurality of nodes, each nonterminal node representing a rhetorical relationship between two fragments of the second plurality of text fragments and each terminal node of the nodes of the discourse tree is associated with one of the fragments of the second plurality of text fragments; and matching each fragment that has a verb to a verb signature; determining whether the search result contains argumentation by providing the communicative discourse tree to a machine learning model trained to detect text comprising argumentation; responsive to determining that the search result contains argumentation, constructing a paragraph from the search result; and providing the paragraph to a user device. ### Claim 2 (depends on 1) The method of claim 1, further comprising training the machine learning model by: accessing a set of training data comprising a training pair, the training pair comprising a first communicative discourse tree that represents text comprising argumentation and a second communicative discourse tree that represents text without argumentation; and providing one of the training pairs to the machine learning model; receiving, from the machine learning model, a determined presence of argumentation; calculating a loss function by calculating a difference between the determined presence of argumentation and an expected presence of argumentation; and adjusting internal parameters of the machine learning model to minimize the loss function. ### Claim 3 (depends on 1) The method of claim 1, wherein obtaining a search result comprises: identifying an entity within the first plurality of text fragments; and providing the entity to the search engine. ### Claim 4 (depends on 3) The method of claim 3, wherein identifying an entity further comprises: constructing a first syntactic parse tree from the first plurality of text fragments; and identifying the entity within the first syntactic parse tree. ### Claim 5 (depends on 4) The method of claim 4, wherein identifying a noun phrase comprises extracting the noun phrase from a node of the first syntactic parse tree. ### Claim 6 (depends on 1) The method of claim 1, further comprising identifying a relevancy metric based on a common entity between a first syntactic parse tree and a second syntactic parse tree, and wherein providing the paragraph to the user device is based on a determination that the relevancy metric is greater than a threshold. ### Claim 7 (depends on 6) The method of claim 6, wherein computing the relevancy metric comprises applying an additional a machine learning model to the first syntactic parse tree and the second syntactic parse tree. ### Claim 8 A system comprising: a non-transitory computer-readable medium storing computer-executable program instructions; and a processing device communicatively coupled to the non-transitory computer-readable medium for executing the computer-executable program instructions, wherein executing the computer-executable program instructions configures the processing device to perform operations comprising: accessing a seed sentence comprising a first plurality of text fragments; obtaining a search result by providing, to a search engine, at least one of the first plurality of text fragments, wherein the search result comprises a second plurality of text fragments; generating, from the search result, a communicative discourse tree, wherein generating the communicative discourse tree comprises: creating a discourse tree from the second plurality of text fragments, wherein the discourse tree comprises a plurality of nodes, each nonterminal node representing a rhetorical relationship between two fragments of the second plurality of text fragments and each terminal node of the nodes of the discourse tree is associated with one of the fragments of the second plurality of text fragments; and matching each fragment that has a verb to a verb signature; determining whether the search result contains argumentation by providing the communicative discourse tree to a machine learning model trained to detect text comprising argumentation; responsive to determining that the search result contains argumentation, constructing a paragraph from the search result; and providing the paragraph to a user device. ### Claim 9 (depends on 8) The system of claim 8, wherein executing the computer-executable program instructions configures the processing device to perform operations comprising training the machine learning model by: accessing a set of training data comprising a training pair, the training pair comprising a first communicative discourse tree that represents text comprising argumentation and a second communicative discourse tree that represents text without argumentation; and providing one of the training pairs to the machine learning model; receiving, from the machine learning model, a determined presence of argumentation; calculating a loss function by calculating a difference between the determined presence of argumentation and an expected presence of argumentation; and adjusting internal parameters of the machine learning model to minimize the loss function. ### Claim 10 (depends on 8) The system of claim 8, wherein obtaining a search result comprises: identifying an entity within the first plurality of text fragments; and providing the entity to the search engine. ### Claim 11 (depends on 10) The system of claim 10, wherein identifying an entity further comprises: constructing a first syntactic parse tree from the first plurality of text fragments; and identifying the entity within the syntactic parse tree. ### Claim 12 (depends on 11) The system of claim 11, wherein identifying a noun phrase comprises extracting the noun phrase from a node of the first syntactic parse tree. ### Claim 13 (depends on 8) The system of claim 8, wherein executing the computer-executable program instructions configures the processing device to perform operations comprising: identifying a relevancy metric based on a common entity between a first syntactic parse tree and a second syntactic parse tree and wherein providing the paragraph to the user device is based on a determination that the relevancy metric is greater than a threshold. ### Claim 14 (depends on 13) The system of claim 13, wherein computing the relevancy metric comprises applying an additional a machine learning model to the first syntactic parse tree and the second syntactic parse tree. ### Claim 15 A non-transitory computer-readable storage medium storing computer-executable program instructions, wherein when executed by a processing device, the computer-executable program instructions cause the processing device to perform operations comprising: accessing a seed sentence comprising a first plurality of text fragments; obtaining a search result by providing, to a search engine, at least one of the first plurality of text fragments, wherein the search result comprises a second plurality of text fragments; generating, from the search result, a communicative discourse tree, wherein generating the communicative discourse tree comprises: creating a discourse tree from the second plurality of text fragments, wherein the discourse tree comprises a plurality of nodes, each nonterminal node representing a rhetorical relationship between two fragments of the second plurality of text fragments and each terminal node of the nodes of the discourse tree is associated with one of the fragments of the second plurality of text fragments; and matching each fragment that has a verb to a verb signature; determining whether the search result contains argumentation by providing the communicative discourse tree to a machine learning model trained to detect text comprising argumentation; responsive to determining that the search result contains argumentation, constructing a paragraph from the search result; and providing the paragraph to a user device. ### Claim 16 (depends on 15) The storage medium of claim 15, wherein when executed by a processing device, the computer-executable program instructions cause the processing device to perform operations comprising training the machine learning model by: accessing a set of training data comprising a training pair, the training pair comprising a first communicative discourse tree that represents text comprising argumentation and a second communicative discourse tree that represents text without argumentation; and providing one of the training pairs to the machine learning model; receiving, from the machine learning model, a determined presence of argumentation; calculating a loss function by calculating a difference between the determined presence of argumentation and an expected presence of argumentation; and adjusting internal parameters of the machine learning model to minimize the loss function. ### Claim 17 (depends on 15) The storage medium of claim 15, wherein obtaining a search result comprises: identifying an entity within the first plurality of text fragments; and providing the entity to the search engine. ### Claim 18 (depends on 17) The storage medium of claim 17, wherein identifying an entity further comprises: constructing a syntactic parse tree from the first plurality of text fragments; and identifying the entity within the first syntactic parse tree. ### Claim 19 (depends on 18) The storage medium of claim 18, wherein identifying a noun phrase comprises extracting the noun phrase from a node of the first syntactic parse tree. ### Claim 20 (depends on 15) The storage medium of claim 15, wherein when executed by a processing device, the computer-executable program instructions cause the processing device to perform operations comprising: identifying a relevancy metric based on a common entity between a first syntactic parse tree and a second syntactic parse tree and wherein providing the paragraph to the user device is based on a determination that the relevancy metric is greater than a threshold.