Information processing system

Abstract

The present invention enables an answer desired by a user to be easily acquired without imposing a burden of work needed for question sentence creation or detailed information searching. In the present invention, input of a user information keyword indicating information regarding a user is received from the user. Input of a question keyword is received from the user. A question sentence is generated that includes the user information keyword and as well as the question keyword. An answer sentence responding to the question sentence is generated. An answer keyword corresponding to the question keyword is extracted from the answer sentence. The answer sentence is presented to the user. The answer sentence is presented to the user such that search processing can be performed using the extracted answer keyword as a search term, or information obtained by performing search processing using said answer keyword as a search term is presented to the user.

Description

Information Processing Systems

[0001] The present invention relates to an information processing system.

[0002] (Prior Art)

[0003] Patent document 1 (JP 2022-180186 A) describes an invention in which dialogue messages are sequentially acquired, keywords are sequentially extracted from the dialogue messages, support information is searched for based on the extracted keywords, and the searched support information is displayed on the screen.

[0004] Patent document 2 (JP Patent Publication No. 2024-6799) describes an invention that accepts input of a natural language search question from a user, accepts an instruction to execute a natural language search from the user, obtains information indicating the circumstances under which the instruction to execute a natural language search was given, and processes the question using the obtained information.

[0005] JP 2022-180186 A JP 2024-6799 A

[0006] An object of the present invention is to enable a user to easily obtain a desired answer without imposing a laborious burden on the user for creating a question and searching for detailed information.

[0007] a question keyword receiving unit that receives input of a question keyword from a user; a question sentence generation unit that generates a question sentence including the user information keyword and the question keyword; an answer sentence generation unit that generates an answer sentence corresponding to the question sentence generated by the question sentence generation unit; an answer keyword extraction unit that extracts an answer keyword corresponding to the question keyword from the answer sentence generated by the answer sentence generation unit; an answer sentence presentation unit that presents the answer sentence generated by the answer sentence generation unit to the user; and a search term / search processing result presentation unit that presents the answer keyword extracted by the answer keyword extraction unit as a search term to the user in a searchable state, or presents information obtained by a search process using the answer keyword as a search term to the user.

[0008] A second aspect is an information processing system according to the first aspect, wherein the question sentence generation unit prepares questions associated with question keywords in advance and generates a question sentence associated with the question keyword received by the question keyword receiving unit.

[0009] A third aspect is the information processing system according to the first aspect, wherein the answer keyword is a proper noun corresponding to the question keyword.

[0010] A fourth aspect is an information processing system in which, in the first or third aspect, the question sentence generation unit generates a question sentence that instructs the system to generate an answer sentence by marking an answer keyword with a specific marking, and the answer keyword extraction unit extracts the answer keyword with the specific marking from the answer sentence.

[0011] A fifth aspect is an information processing system in the first aspect, which includes a database that stores specific information associated with answer keywords, and the search term / search processing result presentation unit presents the specific information to the user so that it can be searched for in the database using the answer keywords extracted by the answer keyword extraction unit as search terms, or presents the specific information obtained by searching the database using the answer keywords as search terms to the user.

[0012] A sixth aspect is an information processing system in which, in the fifth aspect, the database stores proper names of matching candidates in correspondence with answer keywords, and the search term / search processing result presentation unit presents to the user the proper names of the matching candidates in a searchable manner in the database using the answer keywords extracted by the answer keyword extraction unit as search terms, or presents to the user the proper names of the matching candidates obtained by searching the database using the answer keywords as search terms.

[0013] The seventh aspect is an information processing system including an answer sentence generation unit that generates an answer sentence in response to a question received from a user, an answer item presentation unit that presents answer items of the answer sentence to the user so that the user can instruct the answer items of the answer sentence to be excluded from the re-answer sentence, a re-question sentence generation unit that generates a re-question sentence so that the answer items instructed to be excluded by the user are excluded from the re-answer sentence, and a re-answer sentence generation unit that generates a re-answer sentence in response to the re-question sentence generated by the re-question sentence generation unit.

[0014] An eighth aspect is an information processing system comprising an answer sentence generation unit that generates an answer sentence in response to a question received from a user, a determination unit that determines to exclude from a re-answer sentence answer items that overlap with answer items of the current answer sentence generated by the answer sentence generation unit and / or answer items of past answer sentences, a re-question sentence generation unit that generates a re-answer sentence so that the answer items determined to be excluded by the determination unit are excluded from the re-answer sentence, and a re-answer sentence generation unit that generates a re-answer sentence in response to the re-question sentence generated by the re-question sentence generation unit.

[0015] A ninth aspect is an information processing system including an answer sentence generation unit that generates an answer sentence in response to a question received from a user, an answer item presentation unit that presents answer items of the answer sentence to the user so that the user can instruct the user to change the answer content for the answer items of the answer sentence, a re-question sentence generation unit that generates a re-answer sentence so as to obtain a re-answer sentence with changed answer content for the answer item for which the user instructed the answer content to be changed, and a re-answer sentence generation unit that generates a re-answer sentence in response to the re-question sentence generated by the re-question sentence generation unit.

[0016] A tenth aspect is an information processing system in which the answer item presentation unit, the re-question sentence generation unit, and the re-answer sentence generation unit repeatedly operate in the same manner for the re-question sentence generated by the re-question sentence generation unit in the seventh aspect, thereby obtaining multiple answer sentences with no overlapping answer items.

[0017] An eleventh aspect is an information processing system in which, in the eighth aspect, the determination unit, the re-question sentence generation unit, and the re-answer sentence generation unit repeatedly operate in the same manner for the re-question sentence generated by the re-question sentence generation unit, thereby obtaining multiple answer sentences with no overlapping answer items.

[0018] The twelfth aspect is an information processing system in which, in the ninth aspect, the answer item presentation unit, the re-question sentence generation unit, and the re-answer sentence generation unit repeatedly operate in the same manner for the re-question sentence generated by the re-question sentence generation unit, thereby obtaining multiple answer sentences with changed answer content.

[0019] A thirteenth aspect is an information processing system comprising an answer sentence generation unit that generates an answer sentence in response to a question received from a user, a storage unit that stores the answer sentences generated by the answer sentence generation unit, a selection unit that selects multiple answer sentences stored in the storage unit, a re-question sentence generation unit that generates a re-answer sentence so as to obtain a re-answer sentence that integrates the answer contents of the multiple answer sentences selected by the selection unit, and a re-answer sentence generation unit that generates a re-answer sentence in response to the re-question sentence generated by the re-question sentence generation unit.

[0020] A fourteenth aspect is a system including: an answer sentence generation unit that generates an answer sentence in response to a question received from a user;

[0021] The information processing system includes an extraction unit that extracts answer portions corresponding to one or more entities from the answer sentences generated by the answer sentence generation unit, and a storage unit that stores the answer portions extracted by the extraction unit for each of the multiple answer sentences generated by the answer sentence generation unit in correspondence with the one or more entities.

[0022] A fifteenth aspect is an information processing system in which, in the fourteenth aspect, the one or more entities are any one or any combination of problems, needs, seeds, products, markets, functions, structures, patents, papers, statistics, and keywords.

[0023] A sixteenth aspect is an information processing system in the fourteenth or fifteenth aspect, comprising a selection unit that selects one or more entities from the one or more entities, a re-question sentence generation unit that generates a re-question sentence so as to obtain a re-answer sentence that integrates answer portions that are associated with the one or more entities selected by the selection unit and stored in the storage unit, and a re-answer sentence generation unit that generates a re-answer sentence in accordance with the re-question sentence generated by the re-question sentence generation unit.

[0024] A seventeenth aspect is an information processing system in the sixteenth aspect, comprising an answer portion presentation unit that presents to a user answer portions that should be integrated into a re-answer sentence from among the answer portions that are associated with one or more entities selected by the selection unit and stored in the storage unit, allowing the user to indicate which answer portions should be integrated into a re-answer sentence; a re-question sentence generation unit that generates a re-question sentence so that the answer portions indicated by the user as answer portions that should be integrated into the re-answer sentence are integrated into the re-answer sentence; and a re-answer sentence generation unit that generates a re-answer sentence in accordance with the re-question sentence generated by the re-question sentence generation unit.

[0025] The 18th aspect is an information processing system in the 14th aspect, wherein the one or more entities are one item or multiple items of a framework, and the information processing system includes a re-question sentence generation unit that generates a re-answer sentence so as to obtain a re-answer sentence that integrates answer portions that are associated with the one item or multiple items of the framework and stored in the storage unit, and a re-answer sentence generation unit that generates a re-question sentence in accordance with the re-question sentence generated by the re-question sentence generation unit.

[0026] According to the first to eighteenth aspects, the user can easily obtain the answer he or she desires without having to take on the burden of creating a question and searching for detailed information.

[0027] FIG. 1 is a diagram illustrating an example of a configuration of an information processing system according to an embodiment. FIG. 2 is a block diagram illustrating an example of hardware configurations of an administrator terminal, a user terminal, and a server according to an embodiment. FIG. 3 is a block diagram illustrating an example of a functional configuration of an information processing system according to an embodiment. FIG. 4 is a diagram illustrating an example of a question reception screen. FIG. 5A is a diagram illustrating an example of an answer presentation screen. FIG. 5B is a diagram illustrating an example of an answer presentation screen. FIG. 6 is a diagram illustrating an example of an answer presentation screen. FIG. 7 is a diagram illustrating an example of an answer presentation screen. FIG. 8 is a diagram illustrating an example of an answer presentation screen. FIG. 9 is a diagram illustrating an example of a selection screen. FIG. 10A is a diagram illustrating an example of an answer presentation screen. FIG. 10B is a diagram illustrating an example of an answer presentation screen. FIG. 11 is a block diagram illustrating an example of a functional configuration of an information processing system according to a second embodiment. FIG. 12 is a diagram illustrating an example of a user presentation screen. FIG. 13 is a diagram illustrating an example of a second user presentation screen. FIG. 14 is a block diagram illustrating an example of a functional configuration of an information processing system according to a third embodiment. Fig. 15 is a diagram illustrating an input screen for inputting specific information. Fig. 16 is a diagram illustrating a question and answer screen for displaying a question sentence and an answer sentence. Fig. 17A is a diagram illustrating a selection screen for selecting to store information of an answer sentence in a storage unit as specific information. Fig. 17B is a diagram illustrating a selection screen for selecting to store information of an answer sentence in a storage unit as specific information. Fig. 17C is a diagram illustrating a selection screen for selecting to store information of an answer sentence in a storage unit as specific information.

[0028] Hereinafter, an embodiment of the present invention will be described with reference to the drawings.

[0029] FIG. 1 is a diagram illustrating an example of the configuration of an information processing system according to the first embodiment.

[0030] The information processing system is composed of an administrator terminal 100, a user terminal 110, a server 200, and a network 300 that connects the terminals 100, 110 and the server 200 so that data can be sent and received between the terminals 100, 110 and the server 200.

[0031] The administrator terminal 100 and the user terminal 110 are information processing devices, such as personal computers, smartphones, etc. The network 300 is the Internet, an intranet, etc. The server 200 is a server device or a virtual server built on a cloud computing service.

[0032] 2 is a block diagram illustrating an example of the hardware configuration of the administrator terminal 100, the user terminal 110, and the server 200 according to an embodiment. As shown in FIG. 2, in the administrator terminal 100, the user terminal 110, and the server 200, a CPU (Central Processing Unit) 11, a ROM (Read Only Memory) 12, a RAM (Random Access Memory) 13, a storage 14, an input device 16, a display device 17, and a communication I / F 18 are interconnected via a system bus 15 so as to be able to communicate with each other. The CPU 11 is a central processing unit that executes various programs and controls each device connected to the system bus 15. That is, the CPU 11 reads programs from the ROM 12 or the storage 14 and executes the programs using the RAM 13 as a working area. The CPU 11 controls each device connected to the system bus 15 and performs various arithmetic processing in accordance with programs recorded in the ROM 12 or the storage 14. The ROM 12 or the storage 14 stores a BIOS (Basic Input / Output System) and an OS (Operating System), which are control programs executed by the CPU 11, as well as computer-readable and executable programs and various necessary data for realizing this embodiment.

[0033] The ROM 12 stores various control programs and various data. The RAM 13 functions as the main memory, work area, etc. of the CPU 11 and temporarily stores programs or data as a work area. The storage 14 is configured with an HDD (Hard Disk Drive) or an SSD (Solid State Drive) and stores various programs including the BIOS and OS, and various data.

[0034] The input device 16 includes a pointing device such as a mouse and a keyboard, and is used to make various inputs.

[0035] The display device 17 is, for example, a liquid crystal display, and displays various information. The display device 17 may be a touch panel type, and function as the input device 16.

[0036] The display screen of the display device 17 displays various display screens 510 to 570B, which will be described later.

[0037] The communication interface 18 is an interface for communicating with other devices, and uses standards such as Ethernet (registered trademark), FDDI, Wi-Fi (registered trademark), etc. The communication interface 18 connects to the network 300 and controls the transmission and reception of data.

[0038] 3 is a block diagram illustrating an example of the functional configuration of the information processing system according to the embodiment. The functional configuration illustrated in FIG. 3 includes a server 200, an administrator terminal 100, and a user terminal 110.

[0039] The information processing system is a system for presenting answers to the user.

[0040] The information processing system includes a user information keyword receiving unit 1001, a question keyword receiving unit 1002, a question sentence generation unit 1003, an answer sentence generation unit 1004, an answer keyword extraction unit 1005, an answer sentence presentation unit 1006, a search term / search processing result presentation unit 1007, a database 1008, an answer item presentation unit 1009, a re-question sentence generation unit 1010, a re-answer sentence generation unit 1011, a determination unit 1012, a storage unit 1013, a selection unit 1014, an extraction unit 1015, and an answer part presentation unit 1016.

[0041] (First Example)

[0042] The user information keyword receiving unit 1001 receives input of a user information keyword indicating information about the user from the user.

[0043] The question keyword receiving unit 1002 receives a question keyword input from the user.

[0044] The question generation unit 1003 generates a question that includes the user information keywords and the question keywords.

[0045] The answer sentence generation unit 1004 generates an answer sentence in response to the question sentence generated by the question sentence generation unit 1003 .

[0046] The answer keyword extraction unit 1005 extracts answer keywords corresponding to the question keywords from the answer sentences generated by the answer sentence generation unit 1004.

[0047] The answer sentence presenting unit 1006 presents the answer sentence generated by the answer sentence generating unit 1004 to the user.

[0048] The search term / search processing result presentation unit 1007 presents the answer keywords extracted by the answer keyword extraction unit 1005 as search terms to the user so that they can be searched for, or presents information obtained by searching the answer keywords as search terms to the user.

[0049] The database 1008 stores specific information associated with the answer keywords. The database 1008 can be configured as, for example, the database 210 of the server 200 shown in FIG.

[0050] FIG. 4 illustrates an example of a question reception screen 510 .

[0051] The user can input user information keywords and question keywords on the question reception screen 510 displayed on the screen of the display device 17 of the user terminal 110, and perform a process to receive the user information keywords and question keywords.

[0052] The question reception screen 510 has a user information keyword selection section 511 for selecting a user information keyword. For example, if the user is a "manufacturing company," the user information keyword is a word that succinctly indicates the technology possessed by the user. The user information keyword is a concept that includes social issues, technologies, etc. in which the user is interested.

[0053] A question keyword selection section 512 for selecting a question keyword is arranged on the question reception screen 510. The question keyword is, for example, a word that succinctly expresses the content of the question.

[0054] The user information keyword selection section 511 and the question keyword selection section 512 may be configured with radio buttons for alternatively selecting a plurality of keywords, or may be configured with check boxes for selecting a plurality of keywords.

[0055] In the user information keyword selection section 511, it is possible to select user information keywords relating to technical fields such as "metal processing," "resin molding," "sensing," and "smartphone application development."

[0056] The question keyword selection section 512 allows the user to select question keywords related to the content of the question, such as "medical equipment," "treatment," "diagnosis," "prevention," "examination," "nursing care and welfare," "home," "catheter," "surgical robot," "flexible endoscope," and "rigid endoscope."

[0057] A user information keyword and a question keyword are selected in the user information keyword selection section 511 and the question keyword selection section 512, respectively, whereby a process of accepting input of a user information keyword and a question keyword from the user is executed.

[0058] Instead of selecting a user information keyword or a question keyword, the user information keyword or the question keyword may be entered as text. The user information keyword or the question keyword may be composed of a single character string or a combination of multiple character strings.

[0059] When the process of accepting the input of the user information keyword and the question keyword is executed, a question sentence including the user information keyword and the question keyword is generated.

[0060] A specific question QL can be generated by preparing a question QL associated with a user information keyword UK and a question keyword QK in advance and applying the received user information keyword UK and question keyword QK to the question QL. Furthermore, a question QL can be generated that instructs the generation of an answer AL by adding a specific marking to the answer keyword, such as a "black circle," "black circle and right-pointing black triangle," or "left-pointing black triangle and black circle."

[0061] Below are some example questions.

[0062] (Question QL1)

[0063] For example, a question QL1 is prepared in advance: "Please tell me the 10 main QKs that use UK technology. Please be sure to put a black circle both before and after the proper noun that corresponds to each QK."

[0064] When the input of the user information keywords UK1 and UK2, "sensing" and "smartphone app development," and the question keyword QK1, "medical equipment," is accepted, the following question QL1 is generated: "Please tell us the top 10 medical devices that use sensing and smartphone app development technologies. Please be sure to put a black circle both before and after the proper noun that corresponds to each medical device."

[0065] When the question QL1 is generated, an answer AL1 is generated in response to the question QL1.

[0066] The content of the question may be varied depending on the user information keyword UK and the question keyword QK.

[0067] Also, answer rank buttons 513 may be provided to allow the user to select the rank of the answer content, such as the level of detail of the answer content of the answer sentence, granularity, number of words, number of answer items, difficulty / easiness of the answer content (for experts, for the general public, etc.), etc. Furthermore, answer role buttons 514 may be provided to allow the user to select the role of the answerer.

[0068] For example, when the answer role button 514 "like a copywriter" is selected as shown by arrow A in Figure 4, a question containing the question "Please answer like a piece of writing written by a copywriter" is generated, and an answer given from the perspective of the role of a copywriter can be obtained.

[0069] Furthermore, when the answer rank button 513 is selected as indicated by the arrow B in FIG. 4, a question QL2 is generated as follows to obtain a more detailed answer compared to the question QL1.

[0070] (Question text QL2)

[0071] "Please give us the top 10 medical devices that use sensing and smartphone app development technologies. For each medical device, please provide its name, main purpose, main location of use, function, structure, and component technology. Please indicate the symbols "black circle and right-pointing black triangle" before the proper noun that corresponds to each medical device, and "left-pointing black triangle and black circle" after the proper noun. Please also provide search keywords to search for detailed information about each medical device. Please indicate the symbols "black square and right-pointing black triangle" before the string that corresponds to the search keyword itself, and "left-pointing black triangle and black square" after the string. If there are multiple search keywords, separate them with a half-width space."

[0072] When the question QL2 is generated, an answer AL2 is generated in response to the question QL2.

[0073] The answer sentences AL (AL1, AL2) can be automatically created by a document creation system using artificial intelligence (AI). For example, the answer sentences AL (AL1, AL2) can be created by natural language processing using deep learning.

[0074] Alternatively, a chatbot may be used to create answer sentences AL (AL1, AL2). In this case, by posting a question sentence QL (QL1, QL2) to the chatbot, the answer sentences AL (AL1, AL2) can be obtained from the chatbot.

[0075] 5A and 5B show examples of answer presentation screens 520A and 520B, respectively.

[0076] As shown in FIG. 5A, an answer sentence AL1 is presented on an answer presentation screen 520A.

[0077] As shown in Figure 5A, answer sentence AL1 is presented to the user in a searchable form using answer keywords AK1 ("smart watch"), AK2 ("smart glasses"), and AK3 ("blood glucose meter") corresponding to question keyword QK1 ("medical equipment") as search terms.

[0078] Such a reply sentence AL1 is generated, for example, through the following processes S1, S2, and S3.

[0079] (Process S1) An answer sentence AL1' is generated in response to a question sentence QL1 by a document creation system or chatbot using artificial intelligence (AI).

[0080] (Process S2) In answer sentence AL1', specific markings, "black circles," are placed before and after answer keywords AK1 ("smartwatch"), AK2 ("smart glasses"), and AK3 ("blood glucose meter") that correspond to question keyword QK1 ("medical equipment"). Therefore, a process is executed to search answer sentence AL1' and detect the specific markings, "black circles." Then, a process is executed to extract the text written in the areas surrounded by the detected specific markings, "black circles," as answer keywords AK1 ("smartwatch"), AK2 ("smart glasses"), and AK3 ("blood glucose meter") that correspond to question keyword QK1 ("medical equipment").

[0081] (Process S3) In order to enable search processing by clicking on answer keywords AK1 ("smartwatch"), AK2 ("smart glasses"), and AK3 ("blood glucose meter"), the display locations HT1, HT2, and HT3 of answer keywords AK1 ("smartwatch"), AK2 ("smart glasses"), and AK3 ("blood glucose meter") are written in HTML (hypertext markup language), and a process is executed to make the search engines, specific databases, specific web pages, specific web sites, and other HTML files of the hyperlinked URLs accessible.

[0082] Here, the database can store combinations of question keywords QK and answer sentences AL, or / and each item of the answer sentences AL, or / and answer keywords AK in a searchable manner.

[0083] As a result, by clicking on the "smartwatch," "smartglasses," or "blood glucose meter" sections on the answer presentation screen 520A, the user can access a search engine or specific database at a hyperlinked URL, and obtain information such as text, images, and videos related to each of the "smartwatch," "smartglasses," and "blood glucose meter."

[0084] In addition, the user may be presented with information searched for using a search engine using the answer keywords AK1 ("smart watch"), AK2 ("smart glasses"), and AK3 ("blood glucose meter") as search terms, or specific information obtained by searching a specific database.

[0085] For example, the answer presentation screen 520A can display information IM1 such as text, images, videos, etc. that indicate various product names, manufacturers, functions, structures, component technologies, production numbers, market sizes, etc. of smartwatches searched for using answer keyword AK1 ("smartwatch") as a search term. Similarly, information IM2 and IM3 such as text, images, videos, etc. can be displayed for answer keywords AK2 ("smart glasses") and AK3 ("blood glucose meter").

[0086] As shown in FIG. 5B, an answer AL2 in response to the question QL2 is presented on the answer presentation screen 520B.

[0087] On the answer presentation screen 520B, the display areas HT11 and HT12 of the answer keywords AK11 ("vital sign monitor") and AK12 ("vital sign monitor smartphone app") corresponding to the question keyword QK1 ("medical equipment") are hyperlinked to present the user with answer sentence AL2 so that it can be searched via a search engine, specific database, etc.

[0088] In addition, the answer presentation screen 520B may display information such as text, images, videos, etc. indicating various product names, manufacturers, functions, structures, component technologies, production numbers, market sizes, etc., searched using the answer keywords AK11 ("vital sign monitor") and AK12 ("vital sign monitor smartphone app") as search terms.

[0089] Although the answer keywords AK are extracted by detecting specific markings such as "black circles" from the answer sentences AL, it is also possible to extract the answer keywords AK without using specific markings. For example, natural language processing may be used to extract the answer keywords AK corresponding to the question keywords QK from the answer sentences.

[0090] Here, the answer keyword AK may be presented to the user as a search term as is so that it can be searched for, or the search results using that search term may be presented to the user. Alternatively, other keywords, sentences, etc. may be added to the answer keyword AK, or it may be converted into an accurate term, a general name, or a statistical classification name, and then presented to the user as a search term so that it can be searched for, or the search results using that search term may be presented to the user.

[0091] The answer keywords AK extracted by the answer keyword extraction unit may be presented to the user as candidate terms for search terms so that they can be stored in a database and searched for. Information obtained by performing a search process using the answer keywords AK as candidate terms for search terms may be presented to the user.

[0092] Furthermore, multiple answer keywords AK may be present in the answer sentence. Multiple answer keywords may be presented to the user as search terms so that they can be searched, or search results using these multiple search terms may be presented to the user.

[0093] Furthermore, multiple answer keywords among multiple answer sentences obtained by the user repeatedly asking questions may be presented to the user as search terms for search processing, or search results using these multiple search terms may be presented to the user.

[0094] Alternatively, answer keywords in the answer sentence finally obtained by the user repeatedly asking a question may be presented to the user as search terms so that they can be searched for, or search results based on these multiple search terms may be presented to the user. Therefore, answer keywords may be composed of one character string or a combination of multiple character strings and presented to the user as search terms.

[0095] In addition, specific databases to be searched may include, for example, a medical device database (one example is the Pharmaceutical Industry Production Statistics Database), a database of medical device manufacturers, a database of patient numbers, a database of medical procedure numbers, a database of manufacturing companies, a needs database, and a seeds database.

[0096] For example, when an answer sentence AL including an answer content SC such as "The solution is..." is generated in response to a question sentence QL including a question keyword QK such as "needs," an answer keyword AK may be extracted from the answer content SC, and search terms to assist the search may be automatically generated. For example, when an answer keyword AK11 such as "vital sign monitor" is extracted, search terms such as "function," "structure," and "statistics" may be automatically generated by natural language processing and added to "vital sign monitor," or preset search terms such as "function," "structure," and "statistics" may be added to "vital sign monitor," and these search terms may be presented to the user, or search results using these search terms may be presented to the user.

[0097] Furthermore, for example, if the answer keyword AK "echo examination device" is extracted, the search term may be converted to "general-purpose ultrasound imaging diagnostic device," which is the classification name in the Pharmaceutical Industry Production Statistics Database, and presented to the user for search processing, or the search results using that search term may be presented to the user.

[0098] Combinations of question keywords and answer sentences, or / and each item of the answer sentences, or / and answer keywords, can be stored in a searchable manner in the database.

[0099] Below, examples of use according to user attributes will be explained.

[0100] (Examples 1 to 9 when the user is a manufacturing company)

[0101] (Example of use 1) When a user inputs a user information keyword UK that indicates the technology that the user owns, the user is presented with an answer sentence AL with the answer keyword AK being "medical device" that uses the user's technology. By accessing the medical device database and performing a search using "medical device" as the search term, the user can obtain information such as the number of units produced, imported, domestically shipped, and exported for "medical device" that uses the user's technology.

[0102] (Usage example 2) When a user inputs a user information keyword UK that indicates the technology that the user owns, the user is presented with an answer sentence AL with the answer keyword AK being "medical equipment" that uses the user's technology. By performing a search using "medical equipment" as the search term in a search engine, the user can obtain text information, image information, and video information about "medical equipment" that uses the user's technology.

[0103] (Usage example 3) When a user enters a user information keyword UK that indicates the technology that the user owns, the user is presented with an answer sentence AL with the answer keyword AK being "medical equipment" that uses the user's technology. This allows the user to access the database of a medical equipment manufacturer and perform a search using "medical equipment" as the search term to obtain information on medical equipment manufacturers that produce "medical equipment" that uses the user's technology. Note that it is also possible to obtain information on medical equipment manufacturers without using a special database.

[0104] (Usage example 4) When a user enters a user information keyword UK that indicates the technology they own, an answer sentence AL is presented to the user with the answer keyword AK being the "disease / site for which the medical device" that uses their own technology. By accessing a database of patient numbers and performing a search using "medical device" and "disease / site for which it is used" as search terms, the user can obtain information on the number of patients with the "disease / site for which the medical device" that uses their own technology is used. Furthermore, by repeating the question, the user can extract "medical device" and "disease / site for which the medical device is used" as answer keywords AK from each answer sentence obtained.

[0105] (Usage example 5) When a user enters a user information keyword UK that indicates the technology that the user owns, the user is presented with an answer sentence AL with the answer keyword AK being the "diseases and areas where the medical devices" that use the user's technology are used for. By performing a search using "medical devices" and "diseases and areas where the medical devices are used" as search terms in a search engine, the user can obtain text information, image information, and video information about the "diseases and areas where the medical devices are used" that use the user's technology. Furthermore, by repeatedly asking the question, the user can extract "medical devices" and "diseases and areas where the medical devices are used" as answer keywords AK from each answer sentence obtained.

[0106] (Usage Example 6) When a user enters a user information keyword UK that indicates the technology they own, they are presented with an answer sentence AL with the answer keyword AK being "medical procedures (diagnosis, treatment, prevention) in which a "medical device" that uses their own technology. By accessing a database of the number of medical procedures and performing a search using "medical device" and "medical procedures (diagnosis, treatment, prevention) in which it is used" as search terms, the user can obtain information on the number of "medical procedures (diagnosis, treatment, prevention) in which a "medical device" that uses their own technology. Furthermore, by repeating the question, the user can extract "medical device" and "medical procedures (diagnosis, treatment, prevention) in which that medical device is used" as answer keywords AK from each answer sentence obtained.

[0107] (Usage Example 7) When a user enters a user information keyword UK that indicates the technology that the user owns, the user is presented with an answer sentence AL with the answer keyword AK being "medical procedures (diagnosis, treatment, prevention) in which a "medical device" that uses the user's technology. By performing a search using the search terms "medical device" and "medical procedures (diagnosis, treatment, prevention) in which the user's technology is used," the user can obtain information on the number of "medical procedures (diagnosis, treatment, prevention) in which a "medical device" that uses the user's technology is used. Furthermore, by repeatedly asking the question, the user can extract "medical device" and "medical procedures (diagnosis, treatment, prevention) in which the medical device is used" as answer keywords AK from each answer sentence obtained.

[0108] (Usage Example 8) When a user enters a user information keyword UK that indicates the technology they own, an answer sentence AL is presented to the user with the answer keyword AK being the "medical department" of the "medical procedure (diagnosis, treatment, prevention)" in which the "medical device" that uses their own technology. By performing a search using the search terms "medical device," "medical procedure (diagnosis, treatment, prevention) in which it is used," and "medical department," the user can obtain information on academic societies and medical institutions in the "medical department" of the "medical procedure (diagnosis, treatment, prevention)" in which the "medical device" that uses their own technology. Note that from each answer sentence obtained by repeating the question, the "medical device," "the medical procedure (diagnosis, treatment, prevention) in which that medical device is used," and "the medical department for that medical procedure (diagnosis, treatment, prevention)" can be extracted as answer keywords AK.

[0109] (Example 9) When a user inputs a user information keyword UK that indicates the technology that the user possesses, the user is presented with an answer sentence AL with the answer keyword AK being "medical equipment" that uses the user's technology. By accessing the needs database and performing a search using "medical equipment" as the search term, the user can obtain needs for "medical equipment" that uses the user's technology, as well as information on the medical professionals who proposed the needs.

[0110] (Examples 10 and 11 when the user is a medical professional)

[0111] (Usage Example 10) When a user inputs a user information keyword UK that indicates the user's medical practice (diagnosis, treatment, prevention), the user is presented with an answer sentence AL in which the answer keyword AK is the "elementary technology" of a medical device that can solve the problem in the user's medical practice (diagnosis, treatment, prevention). By accessing a database of manufacturing companies and performing a search using "elementary technology" as a search term, the user can obtain information on manufacturing companies that possess the "elementary technology" of a medical device that can solve the problem in the user's medical practice (diagnosis, treatment, prevention).

[0112] (Usage Example 11) When a user inputs a user information keyword UK that indicates the user's medical practice (diagnosis, treatment, prevention), the user is presented with an answer sentence AL with the answer keyword AK being "medical equipment" that can solve the problem in the user's medical practice (diagnosis, treatment, prevention). By accessing the database of a medical equipment manufacturer and performing a search using "medical equipment" as the search term, the user can obtain information on medical equipment manufacturers that produce "medical equipment" that can solve the problem in the user's medical practice (diagnosis, treatment, prevention).

[0113] (Example 12: User is a medical device manufacturer)

[0114] (Usage Example 12) When a user inputs a user information keyword UK that indicates a medical procedure (diagnosis, treatment, prevention) in which the medical device manufactured by the user is used, the user is presented with an answer sentence AL in which the answer keyword AK is the "elementary technology" of a medical device that can solve the problem of the medical procedure (diagnosis, treatment, prevention) in which the medical device manufactured by the user is used. By accessing the database of manufacturing companies and performing a search using "elementary technology" as the search term, the user can obtain information on manufacturing companies that possess the "elementary technology" of a medical device that can solve the problem of the medical procedure (diagnosis, treatment, prevention) in which the medical device manufactured by the user is used.

[0115] According to the above-mentioned use example 3, the manufacturing company can perform optimal matching by selecting the medical device manufacturers indicated in the acquired information as matching candidates.

[0116] Similarly, according to Use Example 8, the manufacturing company can perform optimal matching by selecting medical institutions indicated in the acquired information as matching candidates.

[0117] Similarly, according to Use Example 9, the manufacturing company can perform optimal matching by selecting medical professionals indicated in the acquired information as matching candidates.

[0118] Similarly, according to Use Example 10, medical professionals can perform optimal matching by selecting manufacturing companies indicated in the acquired information as matching candidates.

[0119] Similarly, according to Use Example 11, medical personnel can perform optimal matching by selecting medical device manufacturers indicated in the acquired information as matching candidates.

[0120] Similarly, according to Use Example 12, the medical device manufacturer can perform optimal matching by selecting the manufacturing companies indicated in the acquired information as matching candidates.

[0121] (Second Example)

[0122] Next, a second embodiment will be described. The second embodiment can be implemented in combination with other embodiments. In the following description, portions that overlap with other embodiments will be omitted as appropriate.

[0123] The answer sentence generator 1004 generates an answer sentence in response to a question received from a user.

[0124] The answer item presenting unit 1009 presents the answer items of the answer sentence to the user so that the user can instruct the answer items of the answer sentence to be excluded from the reply sentence.

[0125] The re-question generator 1010 generates a re-question so that answer items instructed to be excluded by the user are excluded from the re-answer.

[0126] The re-answer sentence generation unit 1011 generates a re-answer sentence in response to the re-question sentence generated by the re-question sentence generation unit 1010 .

[0127] The answer item presentation unit 1009, the re-question sentence generation unit 1010, and the re-answer sentence generation unit 1011 operate repeatedly in the same manner for the re-question sentence generated by the re-question sentence generation unit 1010, thereby obtaining multiple answer sentences with no overlapping answer items.

[0128] FIG. 6 shows an example of an answer presentation screen 530 .

[0129] As shown in FIG. 6, the answer presentation screen 530 presents an answer sentence AL11.

[0130] The answer presentation screen 530 presents the answer items SC1, SC2, and SC3 of the answer sentence AL11 to the user so that the user can instruct the answer items SC1, SC2, and SC3 of the answer sentence to be excluded from the re-answer sentence cycle AL12.

[0131] The reply sentence AL11 is generated, for example, through the following processes S11 and S12.

[0132] (Process S11) An answer sentence AL11' is generated in response to the question sentence QL11 by a document creation system or chatbot using artificial intelligence (AI).

[0133] (Process S12) A process is executed in which check boxes 531 are associated with the answer items SC1, SC2, and SC3 of the answer sentence AL11'. Then, a process is executed in which the answer sentence AL11, in which the check boxes 531 are associated with the answer items SC1, SC2, and SC3, is displayed on the display screen of the display device 17 of the user terminal 110. The check boxes 531 are provided so that, when checked, the corresponding answer item is designated as an answer item to be excluded from the next answer sentence.

[0134] On the answer presentation screen 530, the user can check the check boxes 531 corresponding to answer items that the user has already known or that the user does not need and that should be excluded from the next answer sentence. For example, the check boxes 531 corresponding to the answer items SC1, SC2, and SC3 of the answer sentence AL11 are checked.

[0135] Next, a follow-up question QL12 is generated.

[0136] The follow-up question QL12 may be generated by the user manually instructing a follow-up question, or may be generated automatically in response to checking the check box 531.

[0137] A follow-up question QL12 is generated so that the answer items designated for exclusion by the user are excluded from the follow-up answer.

[0138] The follow-up question QL12 is generated, for example, through the following processes S13 and S14.

[0139] (Process S13) A process is executed to detect answer items corresponding to the checked check boxes 531, for example, answer items SC1, SC2, and SC3.

[0140] (Process S14) Based on the detected answer items SC1, SC2, and SC3, a follow-up question QL12 is generated with the content "Please propose a solution for... However, please exclude solutions SC1..., SC2..., SC3...."

[0141] A follow-up answer sentence AL12 is generated in response to the follow-up question sentence QL12.

[0142] The re-reply sentence AL12 is generated, for example, through the following processes S15 and S16.

[0143] (Process S15) A document creation system or chatbot using artificial intelligence (AI) generates a follow-up answer sentence AL12' in response to the follow-up question sentence QL12. The follow-up answer sentence AL12' is generated so as not to include the answer items SC1, SC2, and SC3 that the user has instructed to be excluded.

[0144] (Process S16) A process is executed in which the check boxes 531 are associated with the answer items SC4, SC5, and SC6 of the reply sentence AL12'. Then, a process is executed in which the reply sentence AL12 in which the check boxes 531 are associated with the answer items SC4, SC5, and SC6 is displayed on the display screen of the display device 17 of the user terminal 110.

[0145] Similarly, the user can check the check boxes 531 corresponding to the reply items SC4, SC5, and SC6 of the reply sentence AL12 that the user has determined should be excluded.

[0146] The same process is repeated thereafter, and the next reply sentence AL13' is generated so as not to include the answer items SC1, SC2, SC3, SC4, SC5, and SC6 that the user has indicated should be excluded. In other words, the next reply sentence AL13 will include answer items SC7, SC8, and SC9 that do not overlap with the answer items SC1, SC2, SC3, SC4, SC5, and SC6 that have already been presented to the user.

[0147] It is also possible to restore an answer item that has been designated as being excluded and reflect it in the re-answer sentence. For example, when the user designates the removal and restoration of answer item SC1 on the screen presenting the re-answer sentence AL12, a new re-answer sentence AL14 incorporating the answer content of answer item SC1 may be generated. It is also possible to integrate or organize multiple answer items designated as being excluded and then incorporate them into the re-answer sentence.

[0148] According to the second embodiment, a re-answer sentence that does not include the answer item that the user has designated to be excluded is presented to the user. This allows the user to efficiently obtain multiple answer items. The second embodiment is useful when there is a limit to the amount of information in the answer sentence, such as a limit on the number of characters.

[0149] (Third Example)

[0150] Next, a third embodiment will be described. The third embodiment can be implemented in combination with other embodiments. In the following description, parts that overlap with other embodiments will be omitted as appropriate.

[0151] The answer sentence generator 1004 generates an answer sentence in response to a question received from a user.

[0152] The determination unit 1012 determines to exclude from the re-answer sentence predetermined answer items among the answer items of the current answer sentence generated by the answer sentence generation unit 1004 and / or answer items that overlap with answer items of past answer sentences among the answer items of the current answer sentence generated by the answer sentence generation unit 1004. The predetermined answer items to be excluded are answer items that the user already knows or answer items that are unnecessary for the user.

[0153] The re-question generator 1010 generates a re-question so that the answer items determined to be excluded by the determiner 1012 are excluded from the re-answer.

[0154] The re-answer sentence generation unit 1011 generates a re-answer sentence in response to the re-question sentence generated by the re-question sentence generation unit 1010 .

[0155] The determination unit 1012, the re-question sentence generation unit 1010, and the re-answer sentence generation unit 1011 operate repeatedly in the same manner for the re-question sentence generated by the re-question sentence generation unit 1010, thereby obtaining multiple answer sentences with no overlapping answer items.

[0156] In the third embodiment, instead of the user specifying answer items to be excluded in the second embodiment, the determining unit 1012 automatically determines answer items to be excluded.

[0157] Past answer sentences presented to the user can be stored in, for example, the storage unit 1013 .

[0158] When the answer sentence AL11 is generated in the same manner as in the second embodiment, the current answer items SC1, SC2, and SC3 of the answer sentence AL11 are determined to be answer items to be excluded from the next answer sentence. If there are any overlapping answer items by referring to past answer sentences presented to the user, it may be determined that those answer items are to be excluded. It may also be determined that both the current answer items SC1, SC2, and SC3 and the past answer items are to be excluded.

[0159] Next, a follow-up question QL12 is generated according to the answer items SC1, SC2, and SC3 that have been determined to be excluded, with the content "Please propose a solution for... However, please exclude solutions SC1..., SC2..., SC3...."

[0160] A second answer sentence AL12 is generated in response to the second question sentence QL12.

[0161] Similarly, the reply items SC4, SC5, and SC6 of the reply sentence AL12 are determined to be reply items to be excluded from the next reply sentence.

[0162] The same process is repeated thereafter, and the next reply sentence AL13 is generated so as not to include the answer items SC1, SC2, SC3, SC4, SC5, and SC6 that were determined to be excluded. In other words, the next reply sentence AL13 will include answer items SC7, SC8, and SC9 that do not overlap with the answer items SC1, SC2, SC3, SC4, SC5, and SC6 that have already been presented to the user.

[0163] According to the third embodiment, a re-answer sentence that does not overlap answer items is presented to the user. This allows the user to efficiently obtain multiple answer items. The third embodiment is useful when there is a limit to the amount of information in the answer sentence, such as a limit on the number of characters.

[0164] (Fourth Example)

[0165] Next, a fourth embodiment will be described. The fourth embodiment can be implemented in combination with other embodiments. In the following description, parts that overlap with other embodiments will be omitted as appropriate.

[0166] The answer sentence generator 1004 generates an answer sentence in response to a question received from a user.

[0167] The answer item presenting unit 1009 presents the answer items of the answer sentence to the user so that the user can instruct a change in the answer content for the answer items of the answer sentence.

[0168] The re-question generator 1010 generates a re-question so as to obtain a re-answer with the changed answer content for the answer item for which the user has instructed a change in answer content.

[0169] The re-answer sentence generation unit 1011 generates a re-answer sentence in response to the re-question sentence generated by the re-question sentence generation unit 1010 .

[0170] The answer item presentation unit 1009, the re-question sentence generation unit 1010, and the re-answer sentence generation unit 1011 operate repeatedly in the same manner for the re-question sentence generated by the re-question sentence generation unit 1010, thereby obtaining multiple answer sentences with changed answer content.

[0171] FIG. 7 shows an example of the answer presentation screen 540 .

[0172] As shown in FIG. 7, an answer sentence AL21 is presented on the answer presentation screen 540.

[0173] The answer presentation screen 540 presents the answer items SC1, SC2, and SC3 of the answer sentence AL21 to the user so that the user can instruct changes to the answer contents for the answer items SC1, SC2, and SC3 of the answer sentence.

[0174] The reply sentence AL21 is generated, for example, through the following processes S21 and S22.

[0175] (Process S21) An answer sentence AL21' corresponding to the question sentence QL21 is generated by a document creation system or chatbot using artificial intelligence (AI).

[0176] (Process S22) A process is executed in which the pull-down menu 541 is associated with the answer items SC1, SC2, and SC3 of the answer sentence AL21'. Then, a process is executed in which the answer sentence AL21, in which the pull-down menu 541 is associated with the answer items SC1, SC2, and SC3, is displayed on the display screen of the display device 17 of the user terminal 110. The pull-down menu 541 is provided to change the answer contents of the answer items of the next answer sentence to answer contents corresponding to the selected list.

[0177] It is assumed that answer contents indicating "summary" are written in answer items SC1, SC2, and SC3.

[0178] The user can change the content of the answer in the next answer sentence by selecting a list in the pull-down menu 541 on the answer presentation screen 540. Here, the list selected in the pull-down menu 541 is, for example, "Patent Documents" or "Market." The pull-down menu 541 may also have a list called "No Change."

[0179] For example, "Patent Documents" is selected from the pull-down menu 541 corresponding to the answer items SC1, SC2, and SC3 of the answer sentence AL21.

[0180] Next, a follow-up question QL22 is generated.

[0181] The follow-up question QL22 may be generated by the user manually instructing a follow-up question, or may be generated automatically in response to a list selected from the pull-down menu 541.

[0182] A follow-up question QL22 is generated so that the answer content of the answer item is changed to the answer content according to the list selected by the user.

[0183] The follow-up question QL22 is generated, for example, through the following processes S23 and S24.

[0184] (Process S23) A process is executed to detect the answer items SC1, SC2, SC3 selected from the list in the pull-down menu 541 in association with the contents of the selected list ("patent documents").

[0185] (Process S24) Based on the answer items SC1, SC2, and SC3 for which list selection was detected and the contents of the detected list ("patent documents"), a follow-up question QL22 is generated with the content "Please propose a solution for... However, for solutions SC1..., SC2..., SC3..., please change the answer content to an explanation of the related patent documents."

[0186] A second answer sentence AL22 is generated in response to the second question sentence QL22.

[0187] The re-reply sentence AL22 is generated, for example, through the following processes S25 and S26.

[0188] (Process S25) A document creation system or chatbot using artificial intelligence (AI) generates a follow-up answer sentence AL22' in response to the follow-up question sentence QL22. The follow-up answer sentence AL22' is generated to include answer items SC1, SC2, and SC3 that have been changed to the answer content ("patent document") selected by the user.

[0189] (Process S26) A process is executed in which the pull-down menu 541 is associated with the answer items SC1, SC2, and SC3 of the reply sentence AL22'. Then, a process is executed in which the reply sentence AL22 in which the pull-down menu 541 is associated with the answer items SC1, SC2, and SC3 is displayed on the display screen of the display device 17 of the user terminal 110.

[0190] Similarly, the user can select a list ("Market") different from the previous list using the pull-down menu 541 corresponding to the reply items SC1, SC2, and SC3 of the reply message AL22.

[0191] The same process is repeated thereafter, and the reply contents of reply items SC1, SC2, and SC3 of the next reply sentence AL23 are changed to the reply contents selected by the user ("market").

[0192] According to the fourth embodiment, the answer contents of the answer items in the reply message are changed to the answer contents selected by the user and presented to the user. This allows the user to efficiently obtain multiple answer items with changed answer contents. The fourth embodiment is useful when there is a limit to the amount of information in the reply message, such as a limit on the number of characters.

[0193] (Fifth Example)

[0194] The second and fourth embodiments may be combined and implemented, and the third and fourth embodiments may be combined and implemented.

[0195] FIG. 8 illustrates an example of an answer presentation screen 550 .

[0196] As shown in FIG. 8, the answer presentation screen 550 presents an answer sentence A31.

[0197] The answer presentation screen 550 presents the answer items SC1, SC2, and SC3 of the answer sentence AL31 to the user, allowing the user to instruct the user to exclude answer items SC1, SC2, and SC3 of the answer sentence, change the answer content, or add answer items.

[0198] The reply sentence AL31 is generated, for example, through the following processes S31 and S32.

[0199] (Process S31) An answer sentence AL31' corresponding to the question sentence QL31 is generated by a document creation system or chatbot using artificial intelligence (AI).

[0200] (Process S32) A process is executed in which the Remove button 551 and the Change button 552 are associated with the answer items SC1, SC2, and SC3 of the answer sentence AL31', respectively, and the Add button 553 is associated with the answer items SC1, SC2, and SC3. Then, a process is executed in which the answer sentence AL31, in which the Remove button 551 and the Change button 552 are associated with the answer items SC1, SC2, and SC3, and in which the Add button 553 is associated with the answer items SC1, SC2, and SC3, is displayed on the display screen of the display device 17 of the user terminal 110.

[0201] The exclude button 551 is provided for excluding the corresponding answer item from the next answer sentence.

[0202] The change button 552 is provided to change the answer content of the corresponding answer item in the next answer sentence. The change content indicated by the change button 552 is, for example, "change the wording of the answer content."

[0203] The change button 552 may be replaced with a pull-down menu 541 similar to that in the fourth embodiment.

[0204] The add button 553 is provided for adding an answer item to the next answer sentence.

[0205] The user can remove the corresponding answer item from the next answer sentence by pressing the remove button 551 on the answer presentation screen 550 .

[0206] The user can change the answer content of the answer item corresponding to the next answer sentence by pressing the change button 552 on the answer presentation screen 550.

[0207] The user can add an answer item to the next answer sentence by pressing the add button 553 on the answer presentation screen 550.

[0208] Next, a follow-up question QL32 is generated.

[0209] The follow-up question QL32 may be generated by the user manually instructing a follow-up question, or may be generated automatically in response to pressing the remove button 551, the change button 552, or the add button 553.

[0210] The follow-up question QL32 is generated, for example, through the following processes S33A and S34A.

[0211] (Process S33A) A process is executed to detect the answer items SC1, SC2, and SC3 for which the exclude button 551 has been pressed.

[0212] (Process S34A) Based on the answer items SC1, SC2, and SC3 for which it was detected that the exclude button 551 had been pressed, a follow-up question QL32 is generated with the content "Please propose a solution for... However, please exclude solutions SC1..., SC2..., SC3...."

[0213] The follow-up question QL32 is generated, for example, through the following processes S33B and S34B.

[0214] (Process S33B) A process is executed to detect the answer items SC1, SC2, and SC3 for which the change button 552 has been pressed.

[0215] (Process S34B) Based on the answer items SC1, SC2, and SC3 for which it was detected that the change button 552 had been pressed, a follow-up question QL32 is generated with the content "Please propose a solution for... However, please change the wording of solutions SC1..., SC2..., SC3...."

[0216] The follow-up question QL32 is generated, for example, through the following processes S33C and S34C.

[0217] (Process S33C) A process for detecting that the add button 553 has been pressed is executed.

[0218] (Process S34C) When it is detected that the Add button 553 has been pressed, a follow-up question QL32 is generated with the content "Please propose a solution for... However, please add further answer items to each answer item of solutions SC1..., SC2..., SC3...."

[0219] A second answer sentence AL32 is generated in response to the second question sentence QL32.

[0220] The re-reply sentence AL32 is generated, for example, through the following processes S35 and S36.

[0221] (Process S35) A follow-up answer sentence AL32' corresponding to the follow-up question sentence QL32 is generated by a document creation system or chatbot using artificial intelligence (AI).

[0222] When the user presses the exclusion buttons 551 corresponding to the reply items SC1, SC2, and SC3, a reply sentence AL32' is generated that does not include the reply items SC1, SC2, and SC3.

[0223] When the user presses the change button 552 corresponding to the answer items SC1, SC2, and SC3, a re-answer sentence AL32' is generated with the answer contents of the answer items SC1, SC2, and SC3 changed.

[0224] When the user presses the add button 553, a reply sentence AL32' is generated with a new reply item SC4 added.

[0225] (Process S36) A process is executed in which the remove button 551 and the change button 552 are associated with the answer items SC1, SC2, SC3, and SC4 of the re-answer sentence AL32', respectively, and the add button 553 is associated with the answer items SC1, SC2, SC3, and SC4. Then, a process is executed in which the remove button 551 and the change button 552 are associated with the answer items SC1, SC2, SC3, and SC4, respectively, and the add button 553 is associated with the answer items SC1, SC2, SC3, and SC4, respectively, and the answer sentence AL32 is displayed on the display screen of the display device 17 of the user terminal 110.

[0226] The same process is repeated thereafter, and for the answer items of the next reply sentence AL33, answer items can be excluded, the answer content can be changed, or an answer item can be added depending on the user's button operation. Note that a maintain button for maintaining the answer content may be associated with each of the answer items SC1, SC2, SC3, and SC4 and placed on the answer presentation screen 550. When the maintain button is pressed, the answer content of the answer item for which the maintain button was pressed, for example, answer item SC1, is maintained in the next reply sentence.

[0227] (Sixth Example)

[0228] Next, a sixth embodiment will be described. The sixth embodiment can be implemented in combination with other embodiments. In the following description, parts that overlap with other embodiments will be omitted as appropriate.

[0229] The answer sentence generator 1004 generates an answer sentence in response to a question received from a user.

[0230] The storage unit 1013 stores the answer sentence generated by the answer sentence generation unit 1004 .

[0231] The selection unit 1014 selects a plurality of answer sentences stored in the storage unit 1013 .

[0232] The re-question generator 1010 generates a re-question so as to obtain a re-answer sentence that integrates the answer contents of the multiple answer sentences selected by the selector 1014 .

[0233] The re-answer sentence generation unit 1011 generates a re-answer sentence in response to the re-question sentence generated by the re-question sentence generation unit 1010 .

[0234] The following description will be given with reference to the second to fifth embodiments.

[0235] The storage section 1013 stores an answer sentence AL11 in which answer items SC1, SC2, and SC3 are described, an answer sentence AL12 in which answer items SC4, SC5, and SC6 are described, and an answer sentence AL13 in which answer items SC7, SC8, and SC9 are described.

[0236] FIG. 9 illustrates an example of the selection screen 560 .

[0237] The selection screen 560 has a selection section 561 for selecting the answer contents of the answer sentence AL11, the answer sentence AL12, and the answer sentence AL13 stored in the storage section 1013.

[0238] The selection section 561 is provided for generating a re-answer sentence that integrates the selected answer sentence and / or answer items.

[0239] The selection section 561 can be configured with check boxes corresponding to the answer sentences AL11, AL12, and AL13, respectively. The selection section 561 can also be configured with check boxes corresponding to the answer items SC1, SC2, SC3, SC4, SC5, SC6, SC7, SC8, and SC9, respectively.

[0240] Here, it is assumed that answer items SC1, SC4, and SC7 are selected by checking the selection section 561 as check boxes to be integrated.

[0241] The follow-up question QL41 is generated, for example, through the following processes S41 and S42.

[0242] (Process S41) A process is executed to detect the answer sentences and / or answer items corresponding to the checked checkboxes 561, for example, answer item SC1 of answer sentence AL11, answer item SC4 of answer sentence AL12, and answer item SC7 of answer sentence AL13.

[0243] (Process S42) Based on the detected answer items SC1, SC4, and SC7, a follow-up question QL41 with the content "Please summarize solutions SC1, SC4, SC7." is generated.

[0244] A second answer sentence AL41 is generated in response to the second question sentence QL41.

[0245] The reply sentence AL41 is generated, for example, through the following process S43.

[0246] (Process S43) A document creation system or chatbot using artificial intelligence (AI) generates a follow-up answer sentence AL41 in response to the follow-up question sentence QL41. The follow-up answer sentence AL41 is generated to include answer content that integrates the answer items SC1, SC4, and SC7 that the user instructed to be integrated. The follow-up answer sentence AL41 is stored in the storage unit 1013 as the answer sentence to be integrated.

[0247] In addition, when answer sentences AL11 and AL12 are selected by checking the selection section 561 as check boxes to be integrated, a re-reply sentence AL41 is generated that includes an integrated answer content of the answer contents SC1, SC2, and SC3 of answer sentence AL11 and the answer contents SC4, SC5, and SC6 of answer sentence AL12.

[0248] (Seventh Example)

[0249] Next, a seventh embodiment will be described. The seventh embodiment can be implemented in combination with other embodiments. In the following description, portions that overlap with other embodiments will be omitted as appropriate.

[0250] The answer sentence generator 1004 generates an answer sentence in response to a question received from a user.

[0251] The extraction unit 1015 extracts answer portions corresponding to one or more entities from the answer sentence generated by the answer sentence generation unit 1004 .

[0252] The storage unit 1013 stores the answer portion extracted by the extraction unit 1015 for each of the plurality of answer sentences generated by the answer sentence generation unit 1004 in association with one or more entities.

[0253] The one or more entities may be any one or any combination of the following: Problem PB, Needs NE, Seeds SE, Solution, Research and Development, Product PD, Market MK, Function FC, Structure ST, Patent PT, Paper PA, Statistics SA, Person, Company, Organization Name, and Keyword KY.

[0254] The entity or entities may also be an item or items of a framework such as the Lean Canvas or SWOT analysis.

[0255] The selection unit 1014 selects one or more entities from one or more entities.

[0256] The answer portion presentation unit 1016 presents to the user the answer portion that should be integrated into the re-answer sentence, from among the answer portions stored in the storage unit 1013 in association with one or more entities selected in the selection unit 1014, allowing the user to indicate the answer portion.

[0257] The re-question generation unit 1010 generates a re-question so as to obtain a re-answer sentence that integrates the answer portions stored in the storage unit 1013 in association with one or more entities selected by the selection unit 1014.

[0258] Furthermore, the re-question sentence generator 1010 generates a re-question sentence so that answer portions designated by the user as answer portions to be integrated in the re-answer sentence are integrated in the re-answer sentence.

[0259] Furthermore, the re-question generator 1010 generates a re-question so as to obtain a re-answer that integrates answer portions that are associated with one or more items of the framework and stored in the storage unit 1013 .

[0260] The re-answer sentence generation unit 1011 generates a re-answer sentence in response to the re-question sentence generated by the re-question sentence generation unit 1010 .

[0261] Here, it is assumed that a plurality of reply sentences, for example reply sentences AL51, AL52 and AL53, have already been generated.

[0262] The reply content of reply sentence AL51 includes the reply portion "product PD1..." and the reply portion "market MK1...".

[0263] The reply content of reply sentence AL52 includes the reply portion "product PD2..." and the reply portion "market MK2...".

[0264] The reply content of reply sentence AL53 includes the reply portion "product PD3..." and the reply portion "market MK3...".

[0265] The extraction process S51 by the extraction unit 1015 and the storage process S52 in the storage unit 1013 are performed as follows.

[0266] (Process S51) From the answer sentence AL51, answer parts corresponding to pre-set entities, such as problem PB, need NE, seed SE, solution, R&D, product PD, market MK, function FC, structure ST, patent PT, paper PA, statistics SA, person, company, organization name, and keyword KY, are extracted using natural language processing. Answer sentence AL51 includes an answer part related to product PD, such as "Product PD1...". Answer sentence AL51 also includes an answer part related to market MK, such as "Market MK1...".

[0267] Therefore, the answer portion "Product PD1..." corresponding to the entity called product PD is extracted from answer sentence AL51. Similarly, the answer portion "Market MK1..." corresponding to the entity called market MK is extracted from answer sentence AL51.

[0268] Similarly, the answer portion "Product PD2..." corresponding to the entity called product PD is extracted from answer sentence AL52. Similarly, the answer portion "Market MK2..." corresponding to the entity called market MK is extracted from answer sentence AL52.

[0269] Similarly, the answer portion "Product PD3..." corresponding to the entity Product PD is extracted from answer sentence AL53. Similarly, the answer portion "Market MK3..." corresponding to the entity Market MK is extracted from answer sentence AL53.

[0270] (Process S52)

[0271] The extracted answer parts "Product PD1...", "Product PD2...", "Product PD3..." are stored in the storage unit 1013 in association with the entity called product PD.

[0272] Furthermore, the extracted answer portions "Market MK1...", "Market MK2...", "Market MK3..." are stored in the storage unit 1013 in association with the entity "Market MK".

[0273] FIG. 10A illustrates an example of a selection screen 570A.

[0274] The selection screen 570A has a selection section 571 for selecting each entity, such as problem PB, need NE, seed SE, product PD, market MK, function FC, structure ST, patent PT, paper PA, statistics SA, person, company, organization name, and keyword KY.

[0275] The selection unit 571 is provided to generate a reply sentence AL that integrates the reply portions that are associated with the selected entities and stored in the storage unit 1013 .

[0276] The selection section 571 can be composed of check boxes corresponding to each entity, for example, problem PB, need NE, seed SE, product PD, market MK, function FC, structure ST, patent PT, paper PA, statistics SA, person, company, organization name, and keyword KY.

[0277] On the selection screen 570A, the user can check the checkboxes 571 associated with the entities from which the user wishes to obtain integrated response content. For example, the user checks the checkboxes 571 corresponding to the product PD and the market MK.

[0278] Next, a follow-up question QL61 is generated.

[0279] The follow-up question QL61 may be generated by the user manually instructing a follow-up question, or may be generated automatically in response to checking a check box 571.

[0280] A follow-up question sentence QL61 is generated so that the answer portion of the follow-up reply sentence SL62 is integrated with the entities "product PD" and "market MK" that have been instructed to be integrated by the user.

[0281] The follow-up question QL61 is generated, for example, through the following processes S53 and S54.

[0282] (Process S53) A process is executed to detect the entities "Product PD" and "Market MK" corresponding to the checked checkboxes 571.

[0283] (Process S54) Based on the detected entities "Product PD" and "Market MK", a follow-up question QL61 is generated, which reads, "Please propose a solution for... However, please combine the answer parts "Product PD1...", "Product PD2...", and "Product PD3...". Also, please combine the answer parts "Market MK1...", "Market MK2...", and "Market MK3...". The answer parts "Product PD1...", "Product PD2...", and "Product PD3..." stored in storage unit 1013 in association with the entity "Product PD" are retrieved and incorporated into the follow-up question QL61. Also, the answer parts "Market MK1...", "Market MK2...", and "Market MK3..." stored in storage unit 1013 in association with the entity "Market MK" are retrieved and incorporated into the follow-up question QL61.

[0284] A second answer sentence AL61 is generated in response to the second question sentence QL61.

[0285] The reply sentence AL61 is generated, for example, through the following process S55.

[0286] (Process S55) A document creation system or chatbot using artificial intelligence (AI) generates a re-answer sentence AL61 in response to the re-question sentence QL61. The re-answer sentence AL61 includes an answer portion that combines answer portions "Product PD1...," "Product PD2...," and "Product PD3..." that correspond to the entity "Product PD" that the user instructed to be integrated. The re-answer sentence AL61 also includes an answer portion that combines answer portions "Market MK1...," "Market MK2...," and "Market MK3..." that correspond to the entity "Market MK" that the user instructed to be integrated.

[0287] The reply sentence AL61 is displayed on the display screen of the display device 17 of the user terminal 110.

[0288] The user may specify which part of the answer should be integrated into the reply sentence.

[0289] FIG. 10B illustrates an example of selection screen 570B.

[0290] Selection screen 570B displays the answer portion stored in storage unit 1013 in association with the entity selected on selection screen 570A.

[0291] For example, when the entity "Product PD" is selected on the selection screen 570A, the display screen of the display device 17 of the user terminal 110 transitions to the selection screen 570B, and the answer parts "Product PD1...", "Product PD2...", "Product PD3..." stored in the storage unit 1013 in association with the entity "Product PD" are retrieved and displayed on the selection screen 570B.

[0292] The selection screen 570B has a selection section 572 for selecting each of the answer points "PD11...", "PD12...", "PD13..." in the answer section "Product PD1...", each of the answer points "PD21...", "PD22...", "PD23..." in product PD2..., and each of the answer points "PD31...", "PD32...", "PD33..." in product PD3....

[0293] The selection unit 572 is provided to indicate the answer parts to be integrated into the reply sentence AL from the answer parts stored in the storage unit 1013 in association with the selected entity, and to generate a reply sentence AL that integrates the indicated answer parts.

[0294] The selection section 572 can be configured with check boxes corresponding to each answer location, for example, "PD11...", "PD12...", "PD13...", 210925109 "PD21...", "PD22...", "PD23...", "PD31...", "PD32...", "PD33...".

[0295] On the selection screen 570B, the user can check the check boxes 571 associated with the answer locations from which the user wants to obtain integrated answer content. For example, the user checks the check boxes 572 corresponding to the answer locations "PD11...", "PD21...", and "PD31...".

[0296] Next, a follow-up question QL62 is generated.

[0297] The follow-up question QL62 may be generated by the user manually instructing a follow-up question, or may be generated automatically in response to checking a check box 572.

[0298] A re-question sentence QL62 is generated so that the answer portion of the re-answer sentence SL62 is integrated with the answer portions "PD11...", "PD21...", and "PD31..." that have been instructed to be integrated by the user.

[0299] The follow-up question QL62 is generated, for example, through the following processes S61 and S62.

[0300] (Process S61) A process is executed to detect the answer locations "PD11...", "PD21...", "PD31..." associated with the checked checkboxes 572.

[0301] (Process S62) Based on the detected answer parts "PD11...", "PD21...", and "PD31...", a follow-up question QL61 is generated with the content "Please propose a solution for.... However, please combine the answer parts "PD11...", "PD21...", and "PD31...". The answer parts "PD11...", "PD21...", and "PD31..." from the answer parts "Product PD1...", "Product PD2...", and "Product PD3..." stored in the storage unit 1013 in association with the entity "Product PD" are extracted and incorporated into the follow-up question QL61.

[0302] A second answer sentence AL62 is generated in response to the second question sentence QL61.

[0303] The re-reply sentence AL62 is generated, for example, through the following process S62.

[0304] (Process S62) A document creation system or chatbot using artificial intelligence (AI) generates a follow-up answer sentence AL62 in response to the follow-up question sentence QL61. The follow-up answer sentence AL62 includes an answer portion that combines the answer parts "PD11...," "PD21...," and "PD31..." from the answer portions "Product PD1...," "Product PD2...," and "Product PD3..." that correspond to the entity "Product PD" that the user instructed to be integrated.

[0305] The reply sentence AL62 is displayed on the display screen of the display device 17 of the user terminal 110.

[0306] The second embodiment will be described below.

[0307] The explanation overlapping with the first embodiment will be omitted and only the differences will be explained.

[0308] FIG. 11 is a block diagram illustrating an example of the functional configuration of an information processing system according to the second embodiment.

[0309] The functional configuration illustrated in FIG. 11 is provided in the server 200, the administrator terminal 100, and the user terminal 110.

[0310] The information processing system is a system for presenting answers to the user.

[0311] The information processing system is configured to include a question keyword input unit 2001, a question sentence presentation unit 2002 for user presentation, an answer sentence presentation unit 2003 for user presentation, a re-question instruction unit 2004, a re-question sentence generation unit 2005, a re-answer sentence generation unit 2006, a second answer sentence presentation unit 2007 for user presentation, a question sentence generation unit 1003, and an answer sentence generation unit 1004.

[0312] The question keyword input section 2001 is an on-screen selection / input section that allows a user to select or input a question keyword QK, similar to the question keyword selection section 512. For example, the question keyword input section 2001 can be configured as a pull-down menu 2001A that allows a user to select one question keyword QK from a plurality of question keywords QK.

[0313] The user-presentation question presentation unit 2002 presents a user-presentation question UQL constructed by including the question keyword input unit 2001 .

[0314] The user-presented question UQL is a question to be presented to the user. The user-presented question UQL is a sentence containing a question keyword QK, and is, for example, a short sentence that is a simplified version of the question QL generated by the question generation unit 1003.

[0315] The question generation unit 1003 generates a question QL based on the question keywords QK included in the question UQL to be presented to the user, in the same manner as in the first embodiment.

[0316] When the question QL is generated, the answer sentence generation unit 1004 generates an answer sentence AL corresponding to the question QL in the same manner as in the second embodiment.

[0317] The answer sentence presentation unit 2003 presents an answer sentence UAL for presentation to the user, which corresponds to the answer sentence AL generated by the answer sentence generation unit 1004 .

[0318] The answer sentence UAL for presentation to the user is an answer sentence to be presented to the user. The answer sentence UAL for presentation to the user is, for example, a short sentence summarizing the answer sentence AL.

[0319] The re-question instruction section 2004 indicates a re-question sentence in association with the user-presented answer sentence UAL. The re-question instruction section 2004 can be configured as an instruction button 2004B on the screen.

[0320] When the re-question instruction unit 2004 instructs a re-question, the re-question generation unit 2005 generates a re-question QL based on the answer keywords AK and question keywords QK contained in the answer sentence AL.

[0321] The re-answer sentence generator 2006 generates a re-answer sentence AL in response to the re-question sentence QL generated by the re-question sentence generator 2005 .

[0322] The second answer sentence presentation unit 2007 presents the second answer sentence UAL to the user corresponding to the second answer sentence AL to the user. The second answer sentence UAL to be presented to the user is an answer sentence to be presented to the user. The second answer sentence UAL to be presented to the user is, for example, a short sentence summarizing the second answer sentence AL.

[0323] FIG. 12 shows an example of a screen 610 for presentation to a user.

[0324] The screen 610 for presentation to the user is configured to include a question sentence display section 610A for presentation to the user and an answer sentence display section 610B for presentation to the user.

[0325] FIG. 13 illustrates a second user presentation screen 620 .

[0326] The second screen 620 for presentation to the user includes a section 620A for displaying a re-question sentence for presentation to the user and a section 620B for displaying a second answer sentence for presentation to the user.

[0327] The user can input a question keyword QK on the user presentation screen 610 displayed on the screen of the display device 17 of the user terminal 110, and perform processing to accept the question keyword QK.

[0328] The user-presented question display section 610A displays a user-presented question UQL121, which reads, "Please tell me when to use QK122 in QK121!" The user-presented question UQL121 includes a first question keyword QK121 and a second question keyword QK122. The first question keyword QK121 and the second question keyword QK122 can be selected using pull-down menus 2001A and 2001B, respectively.

[0329] For example, one first question keyword QK121 can be selected from "medical equipment industry," "innovative testing and diagnostic devices," "catheter treatment," and so on in the pull-down menu 2001A.

[0330] In addition, one second question keyword QK122 can be selected from "metal processing," "smartphone app," "software," etc. in the pull-down menu 2001B.

[0331] For example, the user can generate a user-presented question UQL121 such as "Tell me where metal processing (QK122) is used in the medical device industry (QK121)!" by selecting from pull-down menus 2001A and 2001B.

[0332] The user-presented question display section 610A displays an answer instruction button 611. When the user-presented question UQL 121 is generated and an answer corresponding to the generated user-presented question UQL 121 is to be acquired, the answer instruction button 611 is pressed.

[0333] When the answer instruction button 611 is pressed, the question UQL 121 to be presented to the user is accepted.

[0334] When the question sentence UQL121 to be presented to the user is received, the question sentence generation unit 1003 generates the question sentence QL121 based on the first question keyword QK121 ("medical device industry") and the second question keyword QK122 ("metal processing") contained in the question sentence UQL to be presented to the user.

[0335] For example, question QL121 is, "Please list 10 medical devices in the medical device industry (QK121) that use the latest technology in metal processing (QK122), and explain how each is useful for diagnosis and treatment."

[0336] As a result, an answer sentence AL120 is generated that includes answer keywords AK121, AK122, AK123, etc., such as "ultra-precision metal skeleton for prosthetic limbs," "implant cleaning solution using ultrasonic cleaning," "variable shape metal endoscope," etc.

[0337] Based on answer sentence AL120, an answer sentence UAL121 for user presentation corresponding to answer keyword AK121 "ultra-precision metal skeleton for prosthetic limbs", an answer sentence UAL122 for user presentation corresponding to answer keyword AK122 "implant cleaning solution using ultrasonic cleaning", an answer sentence UAL123 for user presentation corresponding to answer keyword AK123 "deformable metal endoscope", etc. are generated.

[0338] The answer sentences UAL121, UAL122, UAL123, etc. for presentation to the user are displayed in the answer sentence display section 610B for presentation to the user on the screen 610 for presentation to the user. An example of the answer sentence UAL121 for presentation to the user is, "The metal skeleton that makes up the prosthetic limb is designed and manufactured using ultra-precise metal processing technology to fit extremely precisely to the skeletal structure of the patient. Skeletons with complex shapes are realized by combining high-precision 3D printing technology and machining technology. We provide highly functional and durable prosthetic limbs that allow natural and flexible movement just like a real skeleton."

[0339] An image corresponding to the answer sentence AL120 may be displayed together with the answer sentence AL120, for example, as a pop-up on the user presentation screen 610. Displaying an image as a pop-up is one example, and the image may be displayed by another method.

[0340] On the screen 610 for presentation to the user, an image instruction button 612 is arranged for each of the answer sentences UAL121, UAL122, UAL123, . . . for presentation to the user.

[0341] When the image instruction button 612 is pressed, an image corresponding to the corresponding answer sentence UAL for user presentation is displayed, for example, as a pop-up on the screen for user presentation 610. Displaying an image as a pop-up is one example, and the image may be displayed by another method.

[0342] For example, when an operation of pressing the image instruction button 612 corresponding to the answer sentence for user presentation UAL121 is performed, an image corresponding to the corresponding answer sentence for user presentation UAL121 (an image relating to "ultra-precision metal skeleton for prosthetic limbs") is displayed, for example, as a pop-up on the screen for user presentation 610. Displaying an image as a pop-up is one example, and the image may be displayed by another method.

[0343] In the user-presented answer sentence display section 610B, a re-ask instruction button 2004B is arranged for each of the user-presented answer sentences UAL121, UAL122, UAL123, . . .

[0344] For example, when the re-question instruction button 2004B is pressed, a re-question according to the corresponding user-presented answer sentence UAL is accepted.

[0345] When an operation of pressing the re-question instruction button 2004B corresponding to the answer sentence UAL121 to be presented to the user is performed, a re-question according to the corresponding answer sentence UAL121 to be presented to the user is accepted.

[0346] When a follow-up question is received in response to the answer sentence UAL121 for user presentation, a follow-up question QL131 is generated based on the answer keyword AK121 ("ultra-precision metal skeleton for prosthetic limbs") contained in the corresponding answer sentence AL120, and the corresponding first question keyword QK121 ("medical device industry") and second question keyword QK122 ("metal processing").

[0347] The follow-up question QL131 reads, "Please explain your solution to the needs in the medical device industry for ultra-precision metal skeletons for prosthetic limbs that utilize metal processing in each of the following categories: 'Purpose / Significance,' 'Appearance / Shape,' 'Function / Structure,' 'Method of Use,' 'Persona,' 'Main Users,' 'Unexpected Users,' 'Advantages,' 'Disadvantages,' 'Revenue Model,' 'Method / Subject / Content of Needs Survey,' 'Elemental Technologies Required for Development,' 'Technical Perspectives Possible to Obtain a Patent,' 'Points to Note During Development,' and 'Innovative and Unique Perspectives Included in the Solution.'" It is desirable for the follow-up question generation unit 2005 to generate the follow-up question QL131 so that at least one of the following answer categories is included in the follow-up answer AK131: 'Main Users,' 'Unexpected Users,' and 'Persona.'

[0348] When the follow-up question QL131 is generated, a follow-up question UQL131 to be presented to the user is generated corresponding to the follow-up question QL131. The follow-up question UQL131 to be presented to the user is, "What is the solution to the need for ultra-precision metal skeletons for prosthetic limbs that utilize metal processing?"

[0349] The screen of the display device 17 of the user terminal 110 transitions from the user presentation screen 610 to a second user presentation screen 620 .

[0350] The user-presented follow-up question UQL 131 is displayed in the user-presented follow-up question display section 620A of the second user-presented screen 620.

[0351] When the re-question sentence QL131 is generated, a re-answer sentence AK131 is generated in response to the re-question sentence QL131. When the re-answer sentence AL131 is generated, a second user-presented answer sentence UAL131 is generated in response to the re-answer sentence AL131.

[0352] The second answer sentence for presentation to the user UAL131 is displayed in the second answer sentence for presentation to the user display section 620B of the second screen for presentation to the user 620.

[0353] The second reply sentence UAL131 to be presented to the user is, for example, as follows:

[0354] "Metal processing is utilized to provide a solution to the needs of the medical device industry for ultra-precise metal skeletons for prosthetic limbs."

[0355] Purpose / significance...

[0356] Appearance and shape...

[0357] Function / Structure...

[0358] How to use...

[0359] Persona...

[0360] Main users:

[0361] An unexpected user...

[0362] merit···

[0363] Disadvantages...

[0364] Revenue model...

[0365] Needs survey method, target, content...

[0366] Elemental technologies required for development...

[0367] Technical perspectives that may be patentable...

[0368] Things to keep in mind when developing...

[0369] The innovative and unique perspectives within the solution..."

[0370] On the second user presentation screen 620, an image instruction button 621 is arranged in association with the second user presentation answer sentence UAL131.

[0371] When the image instruction button 621 is pressed, an image corresponding to the corresponding second answer sentence UAL131 for presentation to the user (an image relating to "ultra-precision metal skeleton for prosthetic limbs utilizing metal processing") is displayed, for example, as a pop-up on the second screen 620 for presentation to the user. Displaying an image as a pop-up is one example, and the image may be displayed by another method.

[0372] An instruction to repeat the follow-up question sentences QL131, QL141, . . . may be given to repeatedly generate follow-up answer sentences AL131, AL141, .

[0373] In the second user presentation answer sentence display section 620B of the second user presentation screen 620, a re-question instruction button 2004B is arranged in association with the second user presentation answer sentence UAL131.

[0374] When the operation of pressing the re-question instruction button 2004B is performed, a re-question according to the corresponding second answer sentence for user presentation UAL131 is accepted.

[0375] When a follow-up question is received in response to the second user-presented answer sentence UAL131, a follow-up question sentence QL131 is generated based on the corresponding answer sentence AL120, the answer keywords AK121 ("ultra-precision metal skeleton for prosthetic limbs") and AK131 ("titanium alloy") contained in the follow-up answer sentence AL131, and the corresponding first question keyword QK121 ("medical device industry") and second question keyword QK122 ("metal processing").

[0376] The new follow-up question QL141 reads, "Please explain your solution to the needs in the medical device industry for ultra-precision metal skeletons for prosthetic limbs that utilize metal processing as an alternative answer to follow-up answer AL131, covering each of the following categories: 'Purpose / Significance,' 'Appearance / Shape,' 'Function / Structure,' 'Method of Use,' 'Persona,' 'Main Users,' 'Unexpected Users,' 'Advantages,' 'Disadvantages,' 'Revenue Model,' 'Method / Subject / Content of Needs Survey,' 'Elemental Technologies Necessary for Development,' 'Technical Perspectives That May Result in Patentability,' 'Points to Note During Development,' and 'Innovative and Unique Perspectives Included in the Solution.'"

[0377] That is, when the re-question instruction unit 2004 instructs the re-questioning unit 2005 to repeat the re-questioning sentences QL131, QL141, etc., the re-questioning sentence generation unit 2005 sequentially generates the next re-questioning sentences AK131, AK141, etc., whose contents include the answer keywords AK121, AK131, etc. in the answer sentences AL121, AL131, etc. that have been obtained up to this point.

[0378] The re-reply sentence generating unit 2006 sequentially generates re-reply sentences AL131, AL141, . . . in response to the re-question sentences AK131, AK141, .

[0379] The second reply sentence presentation unit 2007 sequentially presents to the user second reply sentences UAL131, UAL141, . . . to be presented to the user, which correspond to the sequentially generated re-reply sentences AL131, AL141, .

[0380] When the re-question instruction unit 2004 instructs the user to repeat the re-question sentences QL131, QL141, etc., the second answer sentences UAL131, UAL141, etc. for user presentation are displayed sequentially on the display screen of the display device 17 of the user terminal 110 a predetermined number of times in succession.

[0381] The third embodiment will now be described.

[0382] The description overlapping with the first and second embodiments will be omitted and only the differences will be described.

[0383] For example, when using a conversational AI system that uses a large-scale language model (LLM) to obtain specialized answers about a user-specific, unpublished document (such as a development plan), it is difficult to accurately extract and store only the useful information related to the user-specific document. Furthermore, the system requires repeated specialized and accurate question sentences, which requires skill and is inefficient to operate.

[0384] Furthermore, answers from general-purpose models such as large-scale language models (LLMs) have difficulty distinguishing between specialized information related to user-specific documents and other general information. Even if multiple answers are obtained by repeating a question, these answers may not converge on information related to the user-specific documents, and the information may diverge.

[0385] The objective of the third embodiment is to enable accurate and efficient question answering, knowledge supplementation, answer integration, and the like based on a user-specific, unpublicized document (for example, a development plan).

[0386] FIG. 14 is a block diagram illustrating an example of the functional configuration of an information processing system according to the third embodiment.

[0387] The functional configuration illustrated in FIG. 14 is provided in the server 200, the administrator terminal 100, and the user terminal 110.

[0388] The information processing system is a system for presenting answers to the user.

[0389] The information processing system is composed of a storage unit 3001, an answer engine unit 3002, a selection unit 3003, a judgment engine unit 3004, a storage processing unit 3005, a question engine unit 3006, a rewrite engine unit 3007, an extraction unit 3008, and a display unit 3009.

[0390] The storage unit 3001 stores the specific information DIM.

[0391] The answer engine unit 3002 sequentially learns the specific information DIM stored in the storage unit 3001, and generates an answer sentence AL related to the specific information DIM in response to a received question sentence QL.

[0392] The answer engine unit 3002 also generates the answer sentence AL in a specific format, such as a presentation for investors or for fundraising, a project report, an explanatory document for citizens, or a business plan.

[0393] The selection unit 3003 selects to store the information of the reply sentence AL in the storage unit 3001 as the specific information DIM.

[0394] When the selection unit 3003 selects to store the information of the reply sentence AL as the specific information DIM in the storage unit 3001 , the storage processing unit 3005 stores the information of the reply sentence AL in the storage unit 3001 .

[0395] The determination engine unit 3004 determines that part or all of the information in the reply sentence AL is new specific information that is not currently stored in the storage unit 3001 .

[0396] The judgment engine unit 3004 also evaluates some or all of the information in the answer sentence AL and the existing specific information DIM currently stored in the storage unit 3001 according to specific evaluation criteria, and if it is determined that some or all of the information in the answer sentence AL has a higher evaluation than the existing specific information DIM, it decides to store some or all of the information in the answer sentence AL in the storage unit 3001.

[0397] When the determination engine unit 3004 determines that some or all of the information in the answer sentence AL is new specific information DIM that is not currently stored in the storage unit 3001, the storage processing unit 3005 stores some or all of the information in the answer sentence AL in the storage unit 3001.

[0398] In addition, when the storage processing unit 3005 determines that some or all of the information in the answer sentence AL has a higher evaluation than the existing specific information DIM, it replaces the corresponding information in the existing specific information DIM with some or all of the information in the answer sentence AL and stores it in the storage unit 3001.

[0399] The determination engine unit 3004 may be configured so that the user can select the determination result of the determination engine unit 3004. That is, the determination engine unit 3004 determines that part or all of the information in the answer sentence AL is new specific information that is not currently stored in the storage unit 3001, or determines that part or all of the information in the answer sentence AL is highly evaluated according to specific evaluation criteria in comparison with the specific information DIM currently stored in the storage unit 3001.

[0400] The selection unit 3003 selects information of an answer sentence AL to be stored in the storage unit 3001 from information of an answer sentence AL that has been determined by the determination engine unit 3004 to be new specific information or information of an answer sentence AL that has been determined to have a high evaluation according to specific evaluation criteria.

[0401] The storage processing unit 3005 stores the information of the reply sentence AL selected by the selection unit 3003 in the storage unit 3001 as specific information.

[0402] The question engine unit 3006 generates a question QL related to the specific information DIM in accordance with specific question generation criteria so that the question can be accepted by the answer engine unit 3002 .

[0403] A standard question QLF is prepared in advance, and the question engine unit 3006 extracts a specific character string or multiple specific character strings from the specific information DIM stored in the storage unit 3001, and generates a question QL related to the specific information DIM by replacing the character string or multiple specific character strings in the standard question QLF with the extracted specific character string or multiple specific character strings.

[0404] The rewriting engine unit 3007 rewrites the description content of the specific information DIM stored in the storage unit 3001 in accordance with specific rewriting standards.

[0405] The answer engine unit 3002 sequentially learns the specific information DIM rewritten by the rewriting engine unit 3007, and generates an answer sentence AL related to the specific information DIM in response to the received question sentence QL.

[0406] The extraction unit 3008 compares the information of the answer sentence AL generated by the answer engine unit 3002 with the existing specific information DIM stored in the storage unit 3001, and extracts new specific information NDIM from the information of the generated answer sentence AL that is not stored in the storage unit 3001.

[0407] The display unit 3009 displays the new specific information NDIM extracted by the extraction unit 3008 .

[0408] In addition, the extraction unit 3008 compares the information of the answer sentence AL generated by the answer engine unit 3002 with the existing specific information DIM stored in the storage unit 3001, and extracts, from the existing specific information DIM stored in the storage unit 3001, answer sentence-corresponding specific information ODIM corresponding to the information of the generated answer sentence AL, and new specific information NDIM from the information of the generated answer sentence AL that is not stored in the storage unit 3001.

[0409] The display unit 3009 displays the answer-sentence-related specific information ODIM extracted by the extraction unit 3008 and the new specific information NDIM.

[0410] In addition, the extraction unit 3008 compares the information of the answer sentence AL generated by the answer engine unit 3002 with the existing specific information DIM stored in the storage unit 3001, and extracts the answer sentence-corresponding specific information ODIM that corresponds to the information of the generated answer sentence AL from the existing specific information DIM stored in the storage unit 3001.

[0411] The display unit 3009 displays the answer-text-related specific information ODIM extracted by the extraction unit 3008 .

[0412] The answer engine unit 3002 is composed of artificial intelligence, a virtual assistant, or a chatbot.

[0413] The question engine unit 3006 is composed of artificial intelligence, a virtual assistant, or a chatbot.

[0414] The judgment engine unit 3004 is composed of artificial intelligence, a virtual assistant, or a chatbot.

[0415] The rewriting engine unit 3007 is composed of artificial intelligence, a virtual assistant, or a chatbot.

[0416] For example, an answer engine unit 3002, a question engine unit 3006, a determination engine unit 3004,

[0417] The rewriting engine unit 3007 can use, for example, a large-scale language model (LLM), a virtual assistant using a large-scale language model (LLM), or a chatbot using a large-scale language model (LLM).

[0418] (Eighth Example)

[0419] In the eighth embodiment, the answer engine 3002, question engine 3006, decision engine 3004, and rewrite engine 3007 are each configured with a large-scale language model (LLM).

[0420] 15 shows an example of an input screen 710 for inputting specific information DIM. The input screen 710 includes a specific information input section 711 and an answer engine selection section 712.

[0421] The specific information input section 711 is configured, for example, as a drag and drop area. The specific information input section 711 is an area where specific information DIM such as PDF files of papers, development plans, etc., Word files, image data, video data, etc. can be uploaded by dragging and dropping, or can be read and stored in the storage section 3001 by selecting it.

[0422] The answer engine selection section 712 is configured with, for example, a plurality of answer engine selection buttons 712 B. By performing an operation to select an answer engine selection button 712 B, an answer to the specific information DIM input in the input section 711 can be obtained.

[0423] 16 shows an example of a question and answer screen 720 that displays a question QL and an answer AL. The question QL is displayed in a question display section 720A, and the answer AL is displayed in an answer display section 720B.

[0424] 17A, 17B, and 17C each show an example of a selection screen 730 for selecting to store the information of the reply sentence AL as the identification information DIM in the storage unit 3001.

[0425] An input screen 710, a question and answer screen 720, and a selection screen 730 are displayed on the display screen of the display device 17 of the user terminal 110.

[0426] The user operates the input section 711 on the question and answer screen 720 to store the "cardiopulmonary resuscitation training system development plan" as specific information DIM301 in the storage section 3001.

[0427] The specific information DIM301 is as follows:

[0428] "Cardiopulmonary resuscitation training system development plan

[0429] 1. Development Overview...

[0430] 2. Development goals...

[0431] 3. System functions...

[0432] 4. Development schedule...

[0433] 5. Required resources...

[0434] 6. Evaluation method...

[0435] User testing: User testing with the general public and medical professionals...

[0436] 7. Future prospects...

[0437] 8. Other...

[0438] By selecting the answer engine selection button 712B associated with the specific information DIM301, the question engine unit 3006, answer engine unit 3002, determination engine unit 3004, and rewrite engine unit 3007 associated with the specific information DIM301 are started.

[0439] The question engine unit 3006 generates a question QL301 such as "Who is the main user of the cardiopulmonary resuscitation training system?" based on the specific information DIM301 stored in the storage unit 3001.

[0440] The question engine unit 3006 generates the question QL301 in accordance with a specific question generation criterion, "Generate a question to acquire useful and new information in order to manufacture and sell a cardiopulmonary resuscitation training system and earn a predetermined amount or more of profit per year." For example, an administrator of the information processing system may provide the question engine unit 3006 with a prompt in advance as the specific question generation criterion, such as "Generate a question to acquire useful and new information in order to manufacture and sell products and services related to the documents stored in the storage unit 3001 and earn a predetermined amount or more of profit per year." Note that the system may be configured so that the user can arbitrarily select or input specific question generation criteria.

[0441] The question engine unit 3006 may also generate the question QL301 based on a fixed question QLF. An example of the fixed question QLF is, "Who is the main user of the device or service DK indicated in the user-specific document DIM301?" The user-specific document keyword DK included in this fixed question QLF is, for example, "cardiopulmonary resuscitation training system," which can be extracted from the user-specific specific information DIM301 (e.g., "cardiopulmonary resuscitation training system development plan"). By replacing the user-specific document keyword DK in the fixed question QLF with the extracted specific character string or specific multiple character strings (e.g., "cardiopulmonary resuscitation training system"), the question QL301 related to the specific information DIM301 can be generated.

[0442] The answer engine unit 3002 learns the specific information DIM 301 stored in the storage unit 3001 .

[0443] The answer engine unit 3002 generates an answer sentence AL301 related to the specific information DIM301 in response to the received question sentence QL301 according to the learning result.

[0444] The determination engine unit 3004 determines to store the information included in the answer sentence AL301 in the storage unit 3001 according to a specific evaluation criterion, for example, "information that is useful and novel for manufacturing and selling a cardiopulmonary resuscitation training system and making a predetermined amount or more of profit per year." For example, an administrator of the information processing system may provide the determination engine unit 3004 with a pre-specified instruction (prompt) such as "Please store in the storage unit 3001 information that is useful and novel in the answer sentence AL and makes a predetermined amount or more of profit per year by manufacturing and selling a product or service related to the user's unique specific information DIM." Note that the system may be configured so that the user can arbitrarily select or input specific evaluation criteria.

[0445] Furthermore, the specific evaluation criterion may be an evaluation criterion that "among the information contained in the answer sentence AL, this is new information relative to the existing specific information DIM stored in the storage unit 3001," or an evaluation criterion that "among the information contained in the answer sentence AL, this is useful information but not new relative to the existing specific information DIM stored in the storage unit 3001."

[0446] When the determination engine unit 3004 determines that part or all of the information in the answer sentence AL301 is new and useful specific information that is not currently stored in the storage unit 3001, the storage processing unit 3005 executes a process of storing part or all of the information NDIM1 in the answer sentence AL301 in the storage unit 3001. The storage processing unit 3005 may also replace corresponding information in the existing specific information DIM301 with part or all of the information in the answer sentence AL301.

[0447] Here, the determination result of the determination engine unit 3004 may be selected by the user, and the storage processing unit 3005 may store the information selected by the user in the storage unit 3001.

[0448] In response to the user's selection process, the selection unit 3003 selects part or all of the information NDIM1 from the answer sentence AL301 that has been determined by the determination engine unit 3004 to be new and useful specific information, as information on the answer sentence AL to be stored in the storage unit 3001. For example, the user can select all of the information NDIM1 of the answer sentence AL301. The storage processing unit 3005 stores all of the information NDIM1 of the answer sentence AL selected by the selection unit 3003 in the storage unit 3001 as specific information.

[0449] The new information NDIM1 stored in the storage unit 3001 is as follows:

[0450] "Main users of the CPR training system

[0451] The main users of the products envisioned in this development plan are as follows:

[0452] 1. Healthcare workers

[0453] - Professionals who may perform CPR in a medical setting, such as doctors, nurses, and paramedics.

[0454] ・Market size: Stable demand is expected from medical institutions, emergency medical institutions, medical education institutions, etc. In particular, with the advancement of medical technology, the need for high-quality training is on the rise.

[0455] 2. Emergency medical professionals...

[0456] 3. General public...

[0457] 4. Medical education institutions...

[0458] 5. Companies...

[0459] Additional information on market size...

[0460] In this way, new information NDIM1 is added to the specific information DIM301 stored in the storage unit 3001.

[0461] Next, the rewrite engine unit 3007 rewrites the description contents of the specific information DIM301 and NDIM1 stored in the storage unit 3001 in accordance with specific rewrite criteria. For example, the specific rewrite criteria may be a prompt given in advance to the rewrite engine unit 3007 by an administrator of the information processing system, such as "Please organize the specific information DIM stored in the storage unit 3001 by item in a specified format so that products and services related to the user's specific information DIM can be manufactured and sold to earn a predetermined amount or more in profit per year." Note that the system may be configured so that the user can arbitrarily select or input specific rewrite criteria.

[0462] The specific information DIM301 stored in the storage unit 3001 is updated to specific information DIM302 for the next question.

[0463] Next, the question engine unit 3006 generates a question QL302, for example, "What is the strategic plan for effectively approaching the main users of the cardiopulmonary resuscitation training system?" based on the specific information DIM302 stored in the storage unit 3001, similarly according to specific question generation criteria or based on a standard question QLF.

[0464] Similarly, the answer engine unit 3002 learns the specific information DIM302 stored in the storage unit 3001, and generates an answer sentence AL302 related to the specific information DIM302 in response to the received question sentence QL302 based on the learning results.

[0465] The determination engine unit 3004 similarly determines, according to a specific evaluation criterion, that new and useful information NDIM2 contained in the reply sentence AL302 be stored in the storage unit 3001.

[0466] Similarly, the storage processing unit 3005 executes a process of storing in the storage unit 3001 information NDIM2 of part or all of the reply sentence AL302 that has been determined by the determination engine unit 3004 to be stored in the storage unit 3001 .

[0467] The new information NDIM2 stored in the storage unit 3001 is as follows:

[0468] A plan to effectively approach your main users

[0469] 1. Identifying target users and clarifying their needs

[0470] - Medical professionals: Maintain advanced CPR skills, comply with the latest guidelines, and strengthen collaboration in team medical care.

[0471] ・Emergency medical professionals:

[0472] ・General public: ...

[0473] - Companies: ...improving employee safety awareness, establishing an emergency response system, and complying with legal obligations.

[0474] 2. Approach Strategy

[0475] Healthcare workers:

[0476] Providing information to academic societies, research groups, and professional journals

[0477] Providing demonstrations and trials to medical institutions

[0478] Customized training program development for professionals

[0479] ・Emergency medical professionals:

[0480] ・General public:...

[0481] ・Medical education institutions:

[0482] 3. Promotion strategy...

[0483] 4. Evaluation method...

[0484] 5. Budget planning...

[0485] In this way, new information NDIM2 is added to the specific information DIM302 stored in the storage unit 3001.

[0486] Next, the rewriting engine unit 3007 rewrites the description contents of the specific information DIM302 and NDIM2 stored in the storage unit 3001 in accordance with specific rewriting standards.

[0487] The specific information DIM302 stored in the storage unit 3001 is updated to specific information DIM303 for the next question.

[0488] Next, the question engine unit 3006 generates a question QL303 based on the specific information DIM303 stored in the storage unit 3001, similarly according to specific question generation criteria, or based on a standard question QLF, such as "What is the supplementary information about medical professionals: maintaining advanced CPR skills, complying with the latest guidelines, and strengthening collaboration in team medical care?"

[0489] Similarly, the answer engine unit 3002 learns the specific information DIM303 stored in the storage unit 3001, and generates an answer sentence AL303 related to the specific information DIM303 in response to the received question sentence QL303 based on the learning results.

[0490] Similarly, the determination engine unit 3004 determines to store the information NDIM3 included in the reply sentence AL303 in the storage unit 3001 according to a specific evaluation criterion.

[0491] Similarly, the storage processing unit 3005 executes a process of storing in the storage unit 3001 information NDIM3 of part or all of the answer sentence AL303 determined by the determination engine unit 3004 to be stored in the storage unit 3001.

[0492] The new information NDIM3 stored in the storage unit 3001 is as follows:

[0493] Supplementary information for "Healthcare Professionals: Maintaining Advanced CPR Skills, Adhering to the Latest Guidelines, and Strengthening Collaboration in Team Care"

[0494] 1. Maintaining Advanced CPR Skills:

[0495] Simulation-based learning through short, frequent, high-quality CPR sessions that measure and validate competency is critical for healthcare professionals to maintain life-saving skills.

[0496] CPR chest compression and rescue breathing skills decline with lack of use and practice, reducing the overall effectiveness of CPR, so short, regular practice sessions can help improve skills.

[0497] Mobile, advanced simulation stations within the hospital allow healthcare professionals to regularly refresh their CPR skills and extend the deadline for their course completion cards.

[0498] Maintaining advanced CPR skills:

[0499] To maintain their lifesaving skills, healthcare professionals require short, frequent, high-quality CPR sessions that measure and validate competency.

[0500] The Resuscitation Quality Improvement (RQI) program provides a reliable platform for simulation-based mastery learning.

[0501] Short, regular practice sessions (brief / high frequency instruction) have been shown to improve skills.

[0502] 2. Compliance with the latest guidelines:

[0503] 3. Strengthening collaboration in team medical care:

[0504] In this way, new information NDIM3 is added to the specific information DIM303 stored in the storage unit 3001.

[0505] Next, the rewriting engine unit 3007 rewrites the description contents of the specific information DIM303 and NDIM3 stored in the storage unit 3001 in accordance with specific rewriting standards.

[0506] The specific information DIM 303 stored in the storage unit 3001 is updated to specific information DIM 304 for the next question.

[0507] Next, the question engine unit 3006 generates a question QL304 based on the specific information DIM303 stored in the storage unit 3001, similarly according to specific question generation criteria, or based on a standard question QLF, for example, "Please compile the cardiopulmonary resuscitation training system development plan into a development plan for manufacturing and selling the cardiopulmonary resuscitation training system and earning a specified amount of profit per year or more."

[0508] Similarly, the answer engine unit 3002 learns the specific information DIM304 stored in the storage unit 3001, and generates an answer sentence AL304 related to the specific information DIM304 in response to the received question sentence QL304 based on the learning results.

[0509] Response sentence AL304 is as follows: The specific information NDIM 1 added first is broken down and reconstructed into specific information NDIM 11, NDIM 12, NDIM 13, NDIM 14, NDIM 15, NDIM 16, NDIM 17, and NDIM 18 for each item, and these are reflected in items 1 to 8. The specific information NDIM 2 added the second time is broken down and reconstructed into specific information NDIM 21, NDIM 22, NDIM 23, NDIM 24, NDIM 25, NDIM 26, NDIM 27, and NDIM 28 for each item, and these are reflected in items 1 to 8. In addition, the specific information NDIM 3 added for the third time is decomposed and reconstructed into specific information NDIM 31, NDIM 32, NDIM 33, NDIM 34, NDIM 35, NDIM 36, NDIM 37, and NDIM 38 for each item, and reflected in each item 1 to 8.

[0510] "Cardiopulmonary resuscitation training system development plan

[0511] 1. Development Overview

[0512] ...NDIM 11...NDIM21...NDIM31...

[0513] 2. Development Goals

[0514] ...NDIM 12...NDIM22...NDIM32...

[0515] 3. System Functions

[0516] ...NDIM 13...NDIM23...NDIM33...

[0517] 4. Development Schedule

[0518] ...NDIM 14...NDIM24...NDIM34...

[0519] 5. Required Resources

[0520] ...NDIM 15...NDIM25...NDIM35...

[0521] 6. Evaluation Method

[0522] ...NDIM 16...NDIM26...NDIM36...

[0523] 7. Future Outlook

[0524] ...NDIM 17...NDIM27...NDIM37...

[0525] 8. Other

[0526] ...NDIM 18...NDIM28...NDIM38... ”

[0527] The sequentially generated question sentences QL301, QL302, QL303, and QL304 and answer sentences AL301, AL302, AL303, and AL304 are sequentially displayed and presented to the user in the question sentence display section 720A and the answer sentence display section 720B of the question and answer screen 720 on the display device 17 of the user terminal 110 shown in Fig. 16. The number of times the questions and answers are to be repeated (for example, four times) may be given in advance as an instruction (prompt) to the answer engine unit 3002, the question engine unit 3006, the determination engine unit 3004, and the rewriting engine unit 3007 by, for example, an administrator of the information processing system, or the system may be configured so that the user can specify any number of times by selecting or inputting.

[0528] The above describes a case in which the user inputs specific information DIM301 ("Cardiopulmonary Resuscitation Training System Development Plan") into the specific information input section 711 of the input screen 710, and generates a response statement related to the specific information DIM301 in the response engine section 3002. However, it is also possible to input multiple types of specific information DIM301, DIM401 (e.g., "Project A"), and DIM501 (e.g., "Project B") into the specific information input section 711 of the input screen 710, and then press the response engine selection button 712B on the input screen 710 to select specific information DIM301, DIM401, or DIM501, and have the response engine section 3002 similarly generate a response statement related to the selected specific information DIM301, DIM401, or DIM501.

[0529] In this embodiment, the format of the reply sentence generated by the reply engine unit 3002 is the format of a development plan, but the format of the reply sentence generated by the reply engine unit 3002 can be set arbitrarily. For example, the reply sentence may be generated in a specific format such as the format of a presentation material for investors or for fundraising, the format of a project report, the format of explanatory materials for citizens, or the format of a business plan.

[0530] According to the eighth embodiment, in response to input of specific information DIM301 such as a user-specific, unpublicized document (e.g., a development plan), question answering, knowledge supplementation, answer integration, etc. regarding the specific information DIM301 can be performed accurately and efficiently.

[0531] In the eighth embodiment, question answering, knowledge supplementation, response integration, etc. are performed for specific information DIM301, DIM401, DIM501 such as development plans and project-related documents, but this is not limited to this and the system can be applied to any task such as paper peer review, rating check, patent specification writing, etc.

[0532] (Ninth Example)

[0533] The ninth embodiment will be described below. Processing similar to that of the eighth embodiment will be omitted as appropriate, and only different processing will be described.

[0534] In the eighth embodiment, the question engine unit 3006 automatically generates questions QL301, QL302, QL303, and QL304.

[0535] However, the system may be configured so that the user can manually select or input a question regarding the specific information DIM 301. For example, a plurality of standard questions may be prepared so that the user can edit them, and the user may edit the selected standard question and provide the edited question to the answer engine unit 3002 to instruct the answer.

[0536] In the eighth embodiment, the determination engine unit 3004 automatically determines that part or all of the information in the reply sentences AL301, AL302, and AL303 should be stored in the storage unit 3001.

[0537] However, the configuration may be such that the user selects part or all of the information in the answer sentences AL301, AL302, and AL303, and the selected information is stored in the storage unit 3001.

[0538] (Selection processing example 1)

[0539] 17A, the selection screen 730 includes a storage unit storage information display section 730A and a reply message information display section 730B. The user can select any one character string or specific multiple character strings from the reply message AL301 displayed in the reply message information display section 730B and instruct the selection screen 730 to store the selected information in the storage section 3001. For example, by selecting specific multiple character strings NDIN1 from the reply message AL301 displayed in the reply message information display section 730B and pressing the select button 731, the selected information NDIN1 is stored in the storage section 3001.

[0540] At this time, the extraction unit 3008 extracts new specific information NDIM1 from the information of the answer sentence AL301 in relation to the existing specific information DIM301 stored in the storage unit 3001 in order to assist the user in making a selection.

[0541] To assist the user in making a selection, the display unit 3009 displays the extracted new specific information NDIM1 in a hue, saturation, and brightness that can be distinguished from other information.

[0542] For example, the user can select new specific information NDIM1, which is displayed in red (shown hatched in the figure) in the answer message information display section 730B so that it can be distinguished from the background color (white) of the display screen and other information (black), and instruct it to be stored in the storage section 3001.

[0543] (Selection processing example 2)

[0544] Selection process example 2 will be described with reference to selection screen 730 shown in Fig. 17B. Below, explanations that overlap with selection process example 1 will be omitted and only differences will be described.

[0545] To assist the user in making a selection, the extraction unit 3008 extracts, from the existing specific information DIM301 stored in the storage unit 3001, answer sentence-corresponding specific information ODIM1 corresponding to the information of the generated answer sentence AL301, and new specific information NDIM1 from the information of the answer sentence AL301 that is not stored in the storage unit 3001.

[0546] To assist the user in making a selection, the display unit 3009 displays the answer sentence corresponding specific information ODIM1 and the new specific information NDIM1 in a hue, saturation, and brightness that can be distinguished from other information.

[0547] For example, the user can compare the answer sentence corresponding specific information ODIM1, which is displayed in the answer sentence information display section 730B in blue so that it can be distinguished from other information, with the new specific information NDIM1, which is displayed in red so that it can be distinguished from other information, and select some or all of the information of the new specific information NDIM1 and instruct it to be stored in the storage section 3001.

[0548] (Selection processing example 3)

[0549] Selection process example 3 will be described with reference to selection screen 730 shown in Fig. 17C. Below, explanations that overlap with selection process examples 1 and 2 will be omitted and only differences will be described.

[0550] To assist the user in making a selection, the extraction unit 3008 extracts answer sentence-corresponding specific information ODIM1 corresponding to the information of the generated answer sentence AL301 from the existing specific information DIM301 stored in the storage unit 3001.

[0551] To assist the user in making a selection, the display unit 3009 displays the answer sentence corresponding specific information ODIM1 in a hue, saturation, and brightness that can be distinguished from other information.

[0552] For example, the user can select new specific information NDIM1 by referring to the answer message corresponding specific information ODIM1 displayed in yellow, for example, in the answer message information display section 730B so that it can be distinguished from other information, and instruct the storage section 3001 to store this new specific information NDIM1.

[0553] The disclosure of Japanese Patent Application No. 2024-067079, filed on April 17, 2024, is incorporated herein by reference in its entirety.

[0554] DESCRIPTION OF SYMBOLS 1001 User information keyword receiving unit 1002 Question keyword receiving unit 1003 Question sentence generation unit 1004 Answer sentence generation unit 1005 Answer keyword extraction unit 1006 Answer sentence presentation unit 1007 Search term / search processing result presentation unit 1008 Database 1009 Answer item presentation unit 1010 Re-question sentence generation unit 1011 Re-answer sentence generation unit 1012 Determination unit 1013 Storage unit 1014 Selection unit 1015 Extraction unit 1016 Answer part presentation unit 2001 Question keyword input unit 2002 Question sentence presentation unit for user presentation 2003 Answer sentence presentation unit for user presentation 2004 Re-question instruction unit 2005 Re-question sentence generation unit 2006 Re-answer sentence generation unit 2007 Second answer sentence presentation unit for user presentation 3001 Storage unit 3002 Answer engine unit 3003 Selection unit 3004 Determination engine unit 3005 Storage processing unit 3006 Question engine unit 3007 Rewriting engine unit 3008 Extraction unit 3009 Display unit

Claims

1. An information processing system comprising: a question keyword receiving unit that receives input of question keywords from a user; a question sentence generation unit that generates a question sentence based on the question keywords; an answer sentence generation unit that generates an answer sentence corresponding to the question sentence generated by the question sentence generation unit; and an answer keyword extraction unit that extracts answer keywords corresponding to the question keywords from the answer sentences generated by the answer sentence generation unit.

2. An information processing system as described in claim 1, comprising a user information keyword receiving unit that receives input of user information keywords indicating information about the user from the user, and the question generation unit generates a question based on the user information keywords and the question keywords.

3. The information processing system of claim 1, further comprising a search term / search processing result presentation unit that presents the answer keywords extracted by the answer keyword extraction unit to the user as candidate terms for search terms, which can be stored in a database or searched for, and / or presents to the user information obtained by searching the answer keywords as candidate terms for search terms.

4. The information processing system according to claim 1, wherein the question generation unit is configured to prepare questions associated with question keywords in advance and generate questions associated with the question keywords received by the question keyword reception unit.

5. The information processing system according to claim 1, wherein the answer keyword is a single character string or a combination of multiple character strings corresponding to the question keyword.

6. An information processing system as described in claim 1 or 5, wherein the question generation unit generates a question that instructs the user to generate an answer sentence by marking an answer keyword with a specific marking, and the answer keyword extraction unit extracts the answer keyword with the specific marking from the answer sentence.

7. The information processing system according to claim 1, further comprising a database in which combinations of question keywords and answer sentences, or / and each item of the answer sentences, or / and answer keywords are stored in a searchable manner.

8. The information processing system of claim 7, wherein the search term / search processing result presentation unit presents to the user information on matching candidates in a searchable manner using the answer keyword extracted by the answer keyword extraction unit as a search term, or presents to the user information on matching candidates obtained by search processing using the answer keyword as a search term.

9. An information processing system comprising: an answer sentence generation unit that generates an answer sentence in response to a question received from a user; an answer item presentation unit that presents answer items of the answer sentence to the user so that the user can instruct the answer items of the answer sentence to be excluded from the re-answer sentence; a re-question sentence generation unit that generates a re-question sentence so that the answer items instructed to be excluded by the user are excluded from the re-answer sentence; and a re-answer sentence generation unit that generates a re-answer sentence in response to the re-question sentence generated by the re-question sentence generation unit.

10. An information processing system comprising: an answer sentence generation unit that generates an answer sentence in response to a question received from a user; a re-question sentence generation unit that generates a re-question sentence so that answer items in the current answer sentence generated by the answer sentence generation unit that overlap with predetermined answer items and / or answer items in past answer sentences are excluded from the re-answer sentence; and a re-answer sentence generation unit that generates a re-answer sentence in response to the re-question sentence generated by the re-question sentence generation unit.

11. An information processing system comprising: an answer sentence generation unit that generates an answer sentence in response to a question received from a user; an answer item presentation unit that presents answer items of the answer sentence to the user so that the user can instruct the user to change the answer content for the answer items of the answer sentence; a re-question sentence generation unit that generates a re-question sentence so as to obtain a re-answer sentence with changed answer content for the answer item for which the user instructed the change of answer content; and a re-answer sentence generation unit that generates a re-question sentence in response to the re-question sentence generated by the re-question sentence generation unit.

12. The information processing system of claim 9, wherein the answer item presentation unit, the re-question generation unit, and the re-answer generation unit operate repeatedly in the same manner for the re-question generated by the re-question generation unit, thereby obtaining multiple answer sentences with no overlapping answer items.

13. The information processing system of claim 10, wherein the re-question sentence generation unit and the re-answer sentence generation unit repeatedly operate in the same manner for the re-question sentence generated by the re-question sentence generation unit, thereby obtaining multiple answer sentences with no overlapping answer items.

14. The information processing system of claim 11, wherein the answer item presentation unit, the re-question generation unit, and the re-answer generation unit repeatedly operate in the same manner for the re-question generated by the re-question generation unit, thereby obtaining multiple answer sentences with changed answer content.

15. An information processing system comprising: an answer sentence generation unit that generates an answer sentence in response to a question received from a user; a storage unit that stores the answer sentences generated by the answer sentence generation unit; a selection unit that selects multiple answer sentences stored in the storage unit; a re-question sentence generation unit that generates a re-answer sentence so as to obtain a re-answer sentence that integrates the answer contents of the multiple answer sentences selected by the selection unit; and a re-answer sentence generation unit that generates a re-answer sentence in response to the re-question sentence generated by the re-question sentence generation unit.

16. An information processing system comprising: an answer sentence generation unit that generates an answer sentence in response to a question received from a user; an extraction unit that extracts answer portions corresponding to one or more entities from the answer sentences generated by the answer sentence generation unit; and a storage unit that stores the answer portions extracted by the extraction unit for each of the multiple answer sentences generated by the answer sentence generation unit in correspondence with the one or more entities.

17. The information processing system according to claim 16, wherein the one or more entities are any one or any combination of problems, needs, seeds, solutions, research and development, products, markets, functions, structures, patents, papers, statistics, people, companies, organization names, and keywords.

18. An information processing system as described in claim 16 or 17, comprising: a selection unit that selects one or more entities from the one or more entities; a re-question sentence generation unit that generates a re-question sentence so as to obtain a re-answer sentence that integrates answer portions that are associated with the one or more entities selected by the selection unit and stored in the storage unit; and a re-answer sentence generation unit that generates a re-answer sentence in accordance with the re-question sentence generated by the re-question sentence generation unit.

19. An information processing system as described in claim 18, comprising: an answer portion presentation unit that presents to a user answer portions that should be integrated into a re-answer sentence from among the answer portions that are associated with one or more entities selected by the selection unit and stored in the storage unit, allowing the user to indicate which answer portions should be integrated into a re-answer sentence; a re-question sentence generation unit that generates a re-question sentence so that the answer portions indicated by the user as answer portions that should be integrated into the re-answer sentence are integrated into the re-answer sentence; and a re-answer sentence generation unit that generates a re-answer sentence in accordance with the re-question sentence generated by the re-question sentence generation unit.

20. The information processing system of claim 16, wherein the one or more entities are one or more items of a framework.

21. The information processing system of claim 16, wherein the one or more entities are one or more items of a framework, and the information processing system comprises: a re-question sentence generation unit that generates a re-question sentence so as to obtain a re-answer sentence that integrates answer portions that are associated with the one or more items of the framework and stored in the storage unit; and a re-answer sentence generation unit that generates a re-question sentence in accordance with the re-question sentence generated by the re-question sentence generation unit.

22. The information processing system according to claim 1, further comprising: a user-presentation question presentation unit that presents a user-presentation question constructed including a question keyword input unit.

23. An information processing system according to claim 1 or 2, further comprising: a user-presentation answer sentence presentation unit that presents a user-presentation answer sentence corresponding to the answer sentence generated by the answer sentence generation unit.

24. An information processing system as described in claim 23, comprising: a re-question instruction unit that indicates a re-question sentence in association with the answer sentence for presentation to the user; a re-question sentence generation unit that generates a re-question sentence based on an answer keyword and a question keyword when a re-question is instructed by the re-question instruction unit; a re-answer sentence generation unit that generates a re-answer sentence corresponding to the re-question sentence generated by the re-question sentence generation unit; and a second answer sentence for presentation to the user that corresponds to the re-answer sentence.

25. The information processing system of claim 24, wherein, when the re-question instruction unit issues an instruction to repeat a re-question, the re-question generation unit sequentially generates a next re-question that includes an answer keyword from the answer sentences obtained up to this point, the re-answer generation unit sequentially generates re-answer sentences corresponding to the re-question sentences that are sequentially generated, and the second answer sentence presentation unit sequentially presents to the user second answer sentences for presentation that correspond to the sequentially generated re-answer sentences.

26. An information processing system as described in claim 24 or 25, wherein the follow-up question generation unit generates a follow-up question so that the follow-up answer includes at least one of main users, unexpected users, and personas as answer items.

27. An information processing system as described in claim 25, wherein when the re-question instruction unit instructs the user to repeat the re-question sentence, the second answer sentence for presentation to the user is displayed sequentially on the display screen of the user terminal a predetermined number of times in succession.

28. The information processing system according to claim 1, wherein an image corresponding to the answer sentence is displayed on the screen together with the answer sentence.

29. An information processing system according to claim 24 or 25, wherein an image corresponding to the reply sentence is displayed on the screen together with the reply sentence.

30. An information processing system comprising: a storage unit that stores specific information; an answer engine unit that sequentially learns the specific information stored in the storage unit and generates an answer sentence related to the specific information in response to a received question sentence; a selection unit that selects to store the information of the answer sentence in the storage unit as the specific information; and a storage processing unit that stores the information of the answer sentence in the storage unit when the selection unit selects to store the information of the answer sentence in the storage unit as the specific information.

31. An information processing system comprising: a storage unit that stores specific information; an answer engine unit that sequentially learns the specific information stored in the storage unit and generates an answer sentence related to the specific information in response to a received question sentence; a judgment engine unit that determines that some or all of the information in the answer sentence is new specific information that is not currently stored in the storage unit; and a storage processing unit that stores some or all of the information in the answer sentence in the storage unit when the judgment engine unit determines that some or all of the information in the answer sentence is new specific information that is not currently stored in the storage unit.

32. An information processing system comprising: a storage unit that stores specific information; an answer engine unit that sequentially learns the specific information stored in the storage unit and generates an answer sentence related to the specific information in response to a received question sentence; a judgment engine unit that judges some or all of the information in the answer sentence to be highly rated according to specific evaluation criteria; and a storage processing unit that stores some or all of the information in the answer sentence in the storage unit when the judgment engine unit judges that some or all of the information in the answer sentence is highly rated according to the specific evaluation criteria.

33. An information processing system comprising: a storage unit that stores specific information; an answer engine unit that sequentially learns the specific information stored in the storage unit and generates an answer sentence related to the specific information in response to a received question sentence; a judgment engine unit that determines that some or all of the information in the answer sentence is new specific information not currently stored in the storage unit, or that some or all of the information in the answer sentence has a high evaluation according to specific evaluation criteria compared to the specific information currently stored in the storage unit; a selection unit that selects answer sentence information to be stored in the storage unit from new information in answer sentences determined to be new specific information by the judgment engine unit or information in answer sentences determined to have a high evaluation according to the specific evaluation criteria; and a storage processing unit that stores the answer sentence information selected by the selection unit in the storage unit as the specific information.

34. The information processing system described in claim 32, wherein the judgment engine unit evaluates part or all of the information in the answer sentence and existing specific information currently stored in the storage unit according to specific evaluation criteria, and if it is determined that part or all of the information in the answer sentence has a higher evaluation than the existing specific information, it determines that part or all of the information in the answer sentence should be stored in the storage unit; and if it is determined that part or all of the information in the answer sentence has a higher evaluation than the existing specific information, the storage processing unit replaces corresponding information from the existing specific information with part or all of the information in the answer sentence and stores it in the storage unit.

35. An information processing system according to any one of claims 30 to 33, further comprising a question engine unit that generates a question related to the specific information in accordance with specific question generation criteria, the question being receivable by the answer engine unit.

36. An information processing system as described in claim 35, wherein standard questions are prepared in advance, and the question engine unit extracts a specific character string or multiple specific character strings from specific information stored in the storage unit, and generates a question related to the specific information by replacing the character string or multiple character strings in the standard questions with the extracted specific character string or multiple specific character strings.

37. An information processing system as described in any one of claims 30 to 33, comprising a rewriting engine unit that rewrites the description content of specific information stored in the storage unit in accordance with specific rewriting criteria, wherein the answer engine unit sequentially learns the specific information rewritten by the rewriting engine unit and generates an answer sentence related to the specific information in response to a received question sentence.

38. An information processing system as described in claim 30, comprising: an extraction unit that compares the information of the answer sentence generated by the answer engine unit with existing specific information stored in the storage unit and extracts new specific information from the information of the generated answer sentence that is not stored in the storage unit; and a display unit that displays the new specific information extracted by the extraction unit.

39. An information processing system as described in claim 30, comprising: an extraction unit that compares the information of the answer sentence generated by the answer engine unit with existing specific information stored in the storage unit, and extracts, from the existing specific information stored in the storage unit, answer sentence-corresponding specific information that corresponds to the information of the generated answer sentence, and new specific information from the information of the generated answer sentence that is not stored in the storage unit; and a display unit that distinguishes between the answer sentence-corresponding specific information and the new specific information extracted by the extraction unit and displays them.

40. An information processing system as described in claim 30, comprising: an extraction unit that compares the information of the answer sentence generated by the answer engine unit with existing specific information stored in the storage unit, and extracts answer sentence-corresponding specific information from the existing specific information stored in the storage unit that corresponds to the information of the generated answer sentence; and a display unit that displays the answer sentence-corresponding specific information extracted by the extraction unit.

41. An information processing system according to any one of claims 30 to 33, wherein the answer engine unit generates the answer sentence in a specific format.

42. An information processing system according to any one of claims 30 to 33, wherein the answer engine unit is composed of artificial intelligence, a virtual assistant, or a chatbot.

43. An information processing system according to any one of claims 30 to 33, wherein the decision engine unit is composed of artificial intelligence, a virtual assistant, or a chatbot.

44. The information processing system of claim 35, wherein the question engine unit is composed of artificial intelligence, a virtual assistant, or a chatbot.

45. The information processing system of claim 37, wherein the rewriting engine unit is composed of artificial intelligence, a virtual assistant, or a chatbot.

Images