# US 2023/0053344 A1 · Scenario generation apparatus, scenario generation method, and computer-readablerecording medium > Agents: this data is available as JSON from the patentgrep API and MCP server — see https://patentgrep.com/llms.txt - Number: US 2023/0053344 A1 - Document ID: US20230053344A1 - Status: Published application - Kind: A1 - Application number: 17/797,821 - Filed: 2020-02-21 - Published: 2023-02-23 - Earliest claimed: 2020-02-21 - Assignees: NEC Corporation - Applicants: NEC Corporation - Inventors: Ayako HOSHINO - Web page: https://patentgrep.com/patent/US20230053344A1 ## CPC - G06F 40/134: Hyperlinking - G06F 16/3329: Natural language query formulation - G06F 16/9535: Search customisation based on user profiles and personalisation - G06F 40/137: Hierarchical processing, e.g. outlines - G06F 40/14: Tree-structured documents (parsing G06F40/205; validation G06F40/226) - G06F 40/35: Discourse or dialogue representation - H04L 51/02: using automatic reactions or user delegation, e.g. automatic replies or chatbot-generated messages - G06F 40/56: Natural language generation - G06F: ELECTRIC DIGITAL DATA PROCESSING (computer systems based on specific computational models G06N) - H04L: TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION (arrangements common to telegraphic and telephonic communication H04M) ## Abstract A scenario generation apparatus includes: a keyword selection unit that extracts a hierarchical structure from documents that is a basis of a scenarios for use in chatbot and has the hierarchical structure, and selects keywords from the headings in each layer of the extracted hierarchical structure, a link candidate generation unit that generates a first scenario candidate in which a link is set between the keywords and a second scenario candidate in which a link is not set between the keywords, by using documents, an evaluation value calculation unit that calculates an evaluation value for the link for each of the generated first scenario candidate and the second scenario candidate according to a setting rule, and a scenario generation unit that adopts one of the scenario candidates based on the calculated evaluation value and generates the scenario using the adopted scenario candidate. ## Claims ### Claim 1 A scenario generation apparatus comprising: at least one memory storing instructions; and at least one processor configured to execute the instructions to: extract a hierarchical structure from documents that is a basis of a scenarios for use in chatbot and has the hierarchical structure, and select keywords from the headings in each layer of the extracted hierarchical structure, generate a first scenario candidate in which a link is set between the keywords and a second scenario candidate in which a link is not set between the keywords, by using documents, calculate an evaluation value for the link for each of the generated first scenario candidate and the second scenario candidate according to a setting rule, adopt one of the scenario candidates based on the calculated evaluation value and generate the scenario using the adopted scenario candidate. ### Claim 2 (depends on 1) The scenario generation apparatus according to claim 1, wherein further at least one processor configured to execute the instructions to: generate the first scenario candidate and the second scenario candidate according to the number of selected keywords. ### Claim 3 (depends on 1) The scenario generation apparatus according to claim 1, wherein further at least one processor configured to execute the instructions to: generate for each of the selected keywords, the first scenario candidate in which the link is set between the keywords and the second scenario candidate in which the link is not set between the keywords, by using documents, calculate the evaluation value for the link for each of the first scenario candidate generated for each of the keywords and the second scenario candidate generated for each of the keywords, according to the setting rule, adopt one of the scenario candidates based on the calculated evaluation value for each of the keyword and generate the scenario using the adopted scenario candidate for each of the keywords. ### Claim 4 (depends on 1) The scenario generation apparatus according to claim 1, wherein further at least one processor configured to execute the instructions to: if there is a browsing history for the documents, calculate at least one of a length of a path between the headings, the number of options existing in the path between the headings, and the number of characters in the options existing in the path between the headings, in case of applying the browsing history to the first scenario candidate and the second scenario candidate, as the evaluation value, adopt the scenario candidate having the smallest evaluation value. ### Claim 5 (depends on 1) The scenario generation apparatus according to claim 1, wherein further at least one processor configured to execute the instructions to: if there is no browsing history for the documents, set, for each of the first scenario candidate and the second scenario candidate, two groups including the plurality of headings in different branches of the hierarchical structure, calculate, for each of the first scenario candidate and the second scenario candidate, at least one of a length of a path between the groups, the number of options existing in the path between the groups, and the number of characters in the options existing in the path between the groups, as the evaluation value, adopt the scenario candidate having the smallest evaluation value. ### Claim 6 (depends on 1) The scenario generation apparatus according to claim 1, wherein further at least one processor configured to execute the instructions to: generate the first scenario candidate by setting a link between keywords which exist in different branches. ### Claim 7 (depends on 1) The scenario generation apparatus according to claim 1, wherein further at least one processor configured to execute the instructions to: select keywords satisfying a setting condition from the heading. ### Claim 8 A scenario generation method comprising: extracting a hierarchical structure from documents that is a basis of a scenarios for use in chatbot and has the hierarchical structure, and selecting keywords from the headings in each layer of the extracted hierarchical structure, generating a first scenario candidate in which a link is set between the keywords and a second scenario candidate in which a link is not set between the keywords, by using documents, calculating an evaluation value for the link for each of the generated first scenario candidate and the second scenario candidate according to a setting rule, adopting one of the scenario candidates based on the calculated evaluation value and generating the scenario using the adopted scenario candidate. ### Claim 9 (depends on 8) The scenario generation method according to claim 8, wherein, in the generation of scenario candidates, generating the first scenario candidate and the second scenario candidate according to the number of selected keywords. ### Claim 10 (depends on 8) The scenario generation method according to claim 8, wherein, in the generation of scenario candidates, generating, for each of the selected keywords, the first scenario candidate in which the link is set between the keywords and the second scenario candidate in which the link is not set between the keywords, by using documents, in the calculation of the evaluation value, calculating the evaluation value for the link for each of the first scenario candidate generated for each of the keywords and the second scenario candidate generated for each of the keywords, according to the setting rule, in the generation of scenario, adopting one of the scenario candidates based on the calculated evaluation value for each of the keyword and generating the scenario using the adopted scenario candidate for each of the keywords. ### Claim 11 (depends on 8) The scenario generation method according to claim 8, wherein, if there is a browsing history for the documents, in the calculation of the evaluation value, calculating, at least one of a length of a path between the headings, the number of options existing in the path between the headings, and the number of characters in the options existing in the path between the headings, in case of applying the browsing history to the first scenario candidate and the second scenario candidate, as the evaluation value, in the generation of scenario, adopting the scenario candidate having the smallest evaluation value. ### Claim 12 (depends on 8) The scenario generation method according to claim 8, wherein, if there is no browsing history for the documents, in the calculation of the evaluation value, setting, for each of the first scenario candidate and the second scenario candidate, two groups including the plurality of headings in different branches of the hierarchical structure, in the calculation of the evaluation value, calculating, for each of the first scenario candidate and the second scenario candidate, at least one of a length of a path between the groups, the number of options existing in the path between the groups, and the number of characters in the options existing in the path between the groups, as the evaluation value, in the generation of scenario, adopting the scenario candidate having the smallest evaluation value. ### Claim 13 (depends on 8) The scenario generation method according to claim 8, wherein, in the generation of scenario candidates, generating the first scenario candidate by setting a link between keywords which exist in different branches. ### Claim 14 (depends on 8) The scenario generation method according to claim 8, wherein, in the selection of keywords, selecting keywords satisfying a setting condition from the heading. ### Claim 15 A non-transitory computer-readable recording medium that includes a program, the program including instructions that cause a computer to carry out: extracting a hierarchical structure from documents that is a basis of a scenarios for use in chatbot and has the hierarchical structure, and selecting keywords from the headings in each layer of the extracted hierarchical structure, generating a first scenario candidate in which a link is set between the keywords and a second scenario candidate in which a link is not set between the keywords, by using documents, calculating an evaluation value for the link for each of the generated first scenario candidate and the second scenario candidate according to a setting rule, adopting one of the scenario candidates based on the calculated evaluation value and generating the scenario using the adopted scenario candidate. ### Claim 16 (depends on 15) The non-transitory computer-readable recording medium according to claim 15, wherein, in the generation of scenario candidates, generating the first scenario candidate and the second scenario candidate according to the number of selected keywords. ### Claim 17 (depends on 15) The non-transitory computer-readable recording medium according to claim 15, wherein, in the generation of scenario candidates, generating, for each of the selected keywords, the first scenario candidate in which the link is set between the keywords and the second scenario candidate in which the link is not set between the keywords, by using documents, in the calculation of the evaluation value, calculating the evaluation value for the link for each of the first scenario candidate generated for each of the keywords and the second scenario candidate generated for each of the keywords, according to the setting rule, in the generation of scenario, adopting one of the scenario candidates based on the calculated evaluation value for each of the keyword and generating the scenario using the adopted scenario candidate for each of the keywords. ### Claim 18 (depends on 15) The non-transitory computer-readable recording medium according to claim 15, wherein, if there is a browsing history for the documents, in the calculation of the evaluation value, calculating, at least one of a length of a path between the headings, the number of options existing in the path between the headings, and the number of characters in the options existing in the path between the headings, in case of applying the browsing history to the first scenario candidate and the second scenario candidate, as the evaluation value, in the generation of scenario, adopting the scenario candidate having the smallest evaluation value. ### Claim 19 (depends on 15) The non-transitory computer-readable recording medium according to claim 15, wherein, if there is no browsing history for the documents, in the calculation of the evaluation value, setting, for each of the first scenario candidate and the second scenario candidate, two groups including the plurality of headings in different branches of the hierarchical structure, in the calculation of the evaluation value, calculating, for each of the first scenario candidate and the second scenario candidate, at least one of a length of a path between the groups, the number of options existing in the path between the groups, and the number of characters in the options existing in the path between the groups, as the evaluation value. ### Claim 20 (depends on 15) The non-transitory computer-readable recording medium according to claim 15, wherein, in the generation of scenario candidates, generating the first scenario candidate by setting a link between keywords which exist in different branches. ### Claim 21 (canceled)