Information processing system and program

Abstract

To provide an information processing system and a program capable of generating questions according to the contents of answers from a user and digging deeper into acquired information. [Solution] The control means 20 comprises a user terminal communication unit 30, a generation AI system communication unit 31, a prompt generation unit 32, and an information output unit 33. The prompt generation unit 32 generates an instruction prompt including a question generation instruction to generate additional questions based on basic questions and their answers. After the additional questions are answered, the control means 20 generates an instruction prompt including a question generation instruction to generate additional questions based on the questions sent to the user terminal 12 and the answers received from the user terminal 12, and a target information generation instruction to generate target information based on the acquired information and output the target information if the accuracy of the context information is above a certain level. The information output unit 33 outputs the target information if the information from the generation AI system 15 includes target information.

Description

[Technical Field]

[0001] The present invention relates to an information processing system and a program that conducts staged hearings using a generative AI system. [Background technology]

[0002] In recent years, services have been provided that generate sentences interactively with users using generative AI systems that use large-scale language models (LLMs). Many services using such generative AI systems have been provided. For example, Patent Document 1 discloses a system that generates information (prompts) to be input to a generative AI system without user intervention, transmits the prompts to the generative AI system, retrieves the generated information, and provides it to the user. [Prior art documents] [Patent documents]

[0003] [Patent Document 1] Patent No. 7416390 Summary of the Invention [Problem to be solved by the invention]

[0004] When considering an information processing system that uses a generative AI system to organize information and then makes suggestions based on that organized information, the system needs to acquire a wide and deep understanding of the context (background and situation) of the target information. To achieve this, the system needs to conduct step-by-step interviews with users and dig deeper into the acquired information.

[0005] To conduct staged interviews, it is necessary to prepare staged questions tailored to the purpose in advance. However, in order to extract sufficient information from users, it is necessary not only to have a set of questions, but also to change the content of the questions based on the answers from the users. For this reason, it is possible to prepare questions based on the answers, but this would require preparing a huge number of questions, which is not realistic.

[0006] The present invention has been made in consideration of the above-mentioned problems, and aims to provide an information processing system and program that can generate questions based on the answers provided by users and gradually improve the accuracy of the acquired information. [Means for solving the problem]

[0007] In order to solve the above problem, an information processing system according to the present invention acquires context information through questions posed to a user, and performs a process based on the context information. It is information that can be created and serves a specific purpose. An information processing system that outputs target information, comprising a control means and a storage means, wherein the storage means stores basic questions in advance, the control means comprising a user terminal communication unit that transmits and receives information to and from a user terminal used by the user, a generation AI system communication unit that transmits and receives information to and from a generation AI system, a prompt generation unit that generates instruction prompts for the generation AI system, and an information output unit that outputs the target information in a certain format, wherein the user terminal communication unit transmits the basic questions to the user terminal and receives answers corresponding to the basic questions, and the prompt generation unit generates target information based on context information consisting of the basic questions and the answers corresponding to the basic questions. to Generate additional questions using the AI ​​generation system 1st Includes question generation instructions , 1st and generating an instruction prompt, and the generating AI system communication unit 1stThe user terminal communication unit transmits the additional questions to the user terminal and receives answers corresponding to the additional questions. The prompt generation unit receives the answers transmitted to the user terminal. The above foundation Questions and the above additional questions and the answer received from the user terminal. Again Generate additional questions using the AI ​​generation system Second and a target information generation instruction for generating the target information based on the acquired context information and outputting the target information when the accuracy of the context information is equal to or higher than a certain level. Second Generates an instruction prompt, the generation AI system communication unit receives the additional question or the target information generated by sending the second instruction prompt to the generation AI system, or information indicating that the target information is now ready to be output; The information output unit outputs the target information if the information from the generation AI system received by the generation AI system communication unit includes the target information or information indicating that the target information is now available for output.

[0008] In the information processing system according to the present invention, the prompt generation unit 1st Question generation instructions and the second question generation instruction The above additional questions or the above-mentioned additional questions The AI ​​system generates a plurality of options that serve as answers to the question.

[0009] In the information processing system according to the present invention, the prompt generation unit determines that the accuracy of the context information is equal to or greater than a certain level when the number of items included in the target information to be generated is equal to or greater than a certain level in response to the target information generation instruction. and outputting the target information. Includes instructions.

[0010] In the information processing system according to the present invention, the prompt generation unit determines that the accuracy of the context information is equal to or greater than a certain level, regardless of the number of items included in the target information, when the number of additional questions already generated in the target information generation instruction is equal to or greater than a certain level. and outputting the target information. Includes instructions.

[0011] Furthermore, the information processing system according to the present invention comprises: The purpose information is information that suggests to users how to utilize the generative AI system. .

[0012] Furthermore, the information processing system according to the present invention comprises: The target information is RAG data referenced by the RAG (Retrieval-Augmented Generation) system. .

[0013] In order to solve the above problem, the program of the present invention is a program that obtains context information through questions posed to a user, and creates target information, which is information that can be created based on the context information and is in line with a certain purpose, and causes a computer to execute a user terminal communication process that sends and receives information to and from a user terminal used by the user, a generation AI system communication process that sends and receives information to and from a generation AI system, a prompt generation process that generates an instruction prompt for the generation AI system, and an information output process that outputs the target information in a certain format, wherein the user terminal communication process generates a first instruction prompt including a first question generation instruction that causes the generation AI system to generate an additional question based on the context information consisting of basic questions and answers corresponding to the basic questions that have been stored in advance, the prompt generation process generates the first instruction prompt using the basic questions and the answers corresponding to the basic questions as the context information, the generation AI system communication process sends the first instruction prompt to the generation AI system to receive the generated additional questions, the user terminal communication process sends the additional questions to the user terminal to receive the answers corresponding to the additional questions, and the prompt generation process outputs the target information sent to the user terminal. Basic questions and the above additional questionsand the answer received from the user terminal, and a second instruction prompt is generated that includes a second question generation instruction to have the generation AI system generate additional questions based on the context information consisting of the answer received from the user terminal, and a target information generation instruction to generate the target information based on the acquired context information and enable output of the target information if the accuracy of the context information is above a certain level, and in the generation AI system communication processing, the second instruction prompt is sent to the generation AI system to receive the generated additional questions or the target information or information that the target information is now available to be output, and in the information output processing, if the information from the generation AI system received in the generation AI system communication processing includes the target information or information that the target information is now available to be output, the target information is output. [Effects of the Invention]

[0014] According to the information processing system and program of the present invention, additional questions can be generated in response to the user's answers to the questions, and the user can be asked to answer them, so that the accuracy of the acquired information can be gradually improved before generating the target information. [Brief explanation of the drawings]

[0015] [Figure 1] 1 shows a configuration diagram of an information processing system according to an embodiment of the present invention. [Figure 2] 1 shows a detailed configuration diagram of a control means. [Figure 3] 1 is a flowchart in an information processing system. [Figure 4] 10 is an example of a screen displaying basic questions on a user terminal. [Figure 5] 10 is an example of a screen displaying additional questions on a user terminal. DETAILED DESCRIPTION OF THE INVENTION

[0016] Embodiments of the present invention will be described in detail with reference to the drawings. First, a first embodiment of an information processing system 10 will be described. The information processing system 10 is a system that acquires context information through questions posed to a user and creates goal information based on this context information. The goal information of this information processing system 10 is information that suggests ways to utilize a generative AI system for users who do not use or are not making full use of it in their work. To achieve this, the information processing system 10 asks questions related to the user's work, and then asks additional questions based on the answers obtained from the user, thereby gradually enriching the context information and increasing the accuracy of the goal information.

[0017] The information processing system 10 uses a generative AI system 15 equipped with an LLM to generate follow-up questions and target information. An LLM (Large Scale Language Model) is a language model constructed using large amounts of text data and deep learning technology. This enables the generative AI system 15 to perform natural language processing and generate natural, human-like dialogue and sentences. The generative AI system 15 can receive prompts from the information processing system 10 and transmit corresponding AI output information.

[0018] As shown in Fig. 1, an information processing system 10 can be connected to multiple user terminals 12 via a communication network 16. This allows multiple users to log in and use the information processing system 10. Each user terminal 12 has display means 13 such as a monitor. The user terminal 12 is a computer such as a PC, or a mobile information terminal such as a smartphone. However, the user terminal 12 is not limited to these.

[0019] The information processing system 10 can be connected to a generative AI system 15 via a communication network 16. In FIG. 1, the information processing system 10 is connected to one generative AI system 15, but the information processing system 10 may use multiple generative AI systems 15 as needed. The information processing system 10 itself may include a generative AI system 15. In this embodiment, information is exchanged between the information processing system 10 and the generative AI system 15 by sending and receiving information, but if the information processing system 10 includes a generative AI system 15, information is communicated within the system.

[0020] In Figure 1, the information processing system 10 is connected to multiple user terminals 12 and the generation AI system 15 via a single communication network 16, but the information processing system 10 may also be connected to the user terminals 12 and the generation AI system 15 via separate communication networks. For example, the information processing system 10 can be connected to the user terminals 12 via an in-house LAN and to the generation AI system 15 via the Internet.

[0021] The information processing system 10 can be configured with a computer such as a server. The information processing system 10 includes at least a control means 20, a storage means 21, and a communication means 22. The control means 20 can be configured with a processor such as a CPU. The storage means 21 can be configured using, for example, RAM, DRAM, HDD, SSD, or an external cloud or distributed storage. The communication means 22 can be configured with, for example, a NIC or a wireless LAN connection device.

[0022] As shown in Figure 2, the control means 20 has a user terminal communication unit 30 that sends and receives information to and from the user terminal 12, a generation AI system communication unit 31 that sends and receives information to and from the generation AI system 15, a prompt generation unit 32 that generates instruction prompts, an information output unit 33 that outputs target information in a certain format, and a file analysis unit 34 that analyzes the contents of files sent from the user terminal 12.

[0023] The storage means 21 stores basic questions in advance. The basic questions are basic questions that are tailored to the purpose of the information processing system 10. Since the information processing system 10 of this embodiment proposes a method for utilizing a generative AI system related to business, the basic questions ask basic matters related to the user's business.

[0024] The user terminal communication unit 30 of the control means 20 transmits the basic questions and additional questions that are generated later to the user terminal 12, and receives the answers entered by the user. The user terminal communication unit 30 also receives a file containing information transmitted from the user terminal 12.

[0025] The prompt generation unit 32 of the control means 20 generates an instruction prompt that includes a question generation instruction that causes the generation AI system 15 to generate additional questions based on context information obtained from the user, and a target information generation instruction that causes the generation AI system 15 to generate target information when the context information obtained from the user reaches a certain level of accuracy.

[0026] The AI ​​generation system communication unit 31 of the control means 20 sends an instruction prompt to the AI ​​generation system 15 and receives the generated additional questions or objective information.

[0027] The information analysis unit 34 of the control means 20 acquires information by analyzing the contents of the file transmitted from the user terminal 12. The information analysis unit 34 can also scrape and acquire information about a website from the URL of the website transmitted from the user terminal 12.

[0028] Next, a flow in which the information processing system 10 acquires context information and creates objective information will be described. As shown in Fig. 3, the user terminal communication unit 30 reads basic questions from the storage means 21 and transmits them to the user terminal 12 (S1). The basic questions for proposing how to utilize the generative AI system for business are questions for understanding the overall picture of the user's business, such as the department (department name) and an overview of the business, the individual's assigned role, and goals.

[0029] The user inputs answers to the basic questions on the user terminal 12 (S2). As shown in FIG. 4, the display means 13 of the user terminal 12 displays a question display area 40 in which the questions are displayed, an answer input area 41 in which the answers can be input, and an upload area 42 in which a file can be sent to the information processing system 10 by dragging and dropping it. Also, a button 43 for moving to the next question is displayed at the bottom of the display means 13. When the user inputs an answer and presses the button 43, the answer to the question is sent from the user terminal 12 to the information processing system 10. Also, if the user drags and drops a file into the file upload area 42, the file is sent to the information processing system 10.

[0030] The user terminal communication unit 30 receives the answers to the basic questions sent from the user terminal 12 (S3). If a file has been sent from the user terminal 12, the user terminal communication unit 30 also receives the file. This file is analyzed by the information analysis unit 34.

[0031] Next, the prompt generation unit 32 generates an instruction prompt based on the received information (S4). The instruction prompt includes a question generation instruction for generating additional questions and an instruction for generating target information, as described above. The instruction prompt clearly states the purpose of the information to be generated. For example, the instruction prompt may state, "Your purpose is to collect detailed information about the user's business through a series of questions, and ultimately propose a method for utilizing the generation AI in the user's business."

[0032] The question generation instruction in the instruction prompt uses basic questions and their corresponding answers as initial information to instruct the user to create useful follow-up questions for comprehensively understanding the work content and issues based on this initial information. If a file is sent from the user terminal 12, the initial information includes information analyzed by the information analysis unit 34. The question generation instruction also clearly indicates guidelines for creating follow-up questions. The guidelines include, for example, instructions to ask specific and easy-to-answer questions, to use the provided information to ask relevant follow-up questions, to utilize metacognition to ask a variety of questions from multiple perspectives, and to explore the user's work from various angles.

[0033] The question generation instructions also include instructions to create options for answers to the follow-up questions. For this purpose, the question generation instructions include, for example, a statement such as, "For each question, create three possible example answers, which will be representative answers based on the information provided. The example answers should be simple and concise." By generating multiple options corresponding to the follow-up questions, the options are presented to the user when answering the follow-up questions, eliminating the need for text input and making it easier to gather information. Note that when answering the follow-up questions, the user can also enter any answer in text in addition to the generated options.

[0034] The question generation instructions include instructions for the generative AI system 15 to consider the following items: (1) a summary of the current information, (2) information gaps, (3) potential generative AI applications, (4) an evaluation of the progress of information gathering, and (5) a plan for the next questions. The question generation instructions also include instructions for additional guidelines when generating questions. The additional guidelines include instructions on how to ask questions, such as the number of questions to ask and avoiding repeating the same type of question.

[0035] The instruction prompt includes a destination information generation instruction that generates destination information based on the acquired context information and outputs the destination information if the accuracy of the context information is above a certain level. To generate the destination information, the destination information generation instruction includes, for example, an instruction to compile and output important findings obtained from the consideration items once it is determined that sufficient information has been collected to make a proposal. The destination information generation instruction includes an instruction to output the destination information if the generation AI system 15 generates a predetermined number of proposal items based on the consideration instructions in the question generation instruction, assuming that the accuracy of the acquired context information is above a certain level. Furthermore, the destination information generation instruction may include an instruction to output the destination information if the number of additional questions already generated is above a certain level, regardless of the number of proposal items, regarding the accuracy of the context information.

[0036] The instruction prompt instructs the generation AI system 15 to output target information when the context information is accurate to a certain level, so that the generation AI system 15 outputs additional questions when the context information is not accurate enough, and outputs target information when the context information is accurate enough.

[0037] The user may be able to instruct the user terminal 12 to immediately output the target information. In this case, the instruction prompt generated by the prompt generator 32 includes an instruction to the AI ​​system 15 to output the generated target information regardless of the accuracy of the context information.

[0038] Once the instruction prompt is generated, the generation AI system communication unit 31 sends the instruction prompt to the generation AI system 15 (S5) and receives additional questions or objective information generated by the generation AI system 15 (S6).

[0039] Examples of additional questions and options generated by the generation AI system 15 in the information processing system 10 of this embodiment will be given below. (Additional questions) What is the most time-consuming or burdensome task in PM work? (Choices) 1. Create and update project plans and status reports 2. Coordinating communication between teams and creating meeting minutes 3. Risk management and issue documentation, and consideration of countermeasures (Additional questions) When planning a new service, what process do you follow to gather information and come up with ideas? (Choices) 1. Analysis of market research reports and benchmarking of competitor services 2. Conduct internal brainstorming and customer interviews 3.Analysis of industry trends and research of user needs (Additional questions) What types of documents do you typically create when proposing to clients, and what challenges do you face when creating them? (Choices) 1. Proposals and quotes take time to create and require customization 2. Creating technical specifications and preparing explanatory materials is complicated 3. It is difficult to search for and reuse past cases of similar projects

[0040] If the information received in S6 is target information (S7), the information output unit 33 outputs the target information (S8). If the information received in S6 is an additional question (S7), the user terminal communication unit 30 transmits the additional question to the user terminal 12 (S9). The information generated by the generation AI system 15 includes options for the additional question, and the user terminal communication unit 30 transmits information about the options together with the additional question to the user terminal 12.

[0041] As shown in FIG. 5, the display means 13 of the user terminal 12 displays a plurality of options 51 below a question display area 50 in which additional questions are displayed. Also, a button 53 for moving on to the next question is displayed at the bottom of the display means 13. The user can move on to the next additional question by selecting one or more of the options 51, or by entering an arbitrary answer if they select "Other" and pressing the button 53. Information on the option selected by the user or the answer entered by the user is transmitted from the user terminal 12 to the information processing system 10 (S10).

[0042] The user terminal communication unit 30 receives information on the answers to the additional questions from the user terminal 12 (S11). The additional questions and the information on the answers to the additional questions are used as context information, and steps from S4 onwards are repeated. If the generation AI system 15 determines that the context information is sufficiently accurate, the generation AI system 15 transmits target information, which is then output from the information output unit 33 in S8.

[0043] The objective information generated by the generative AI system 15 includes information such as a summary of the context information, identified issues, proposed uses of the generative AI system based on the collected information, effects, and priorities. The information output unit 33 outputs the objective information from the generative AI system 15 in a certain format. In this embodiment, information proposing how to utilize the generative AI system 15 is output, for example, as follows: Proposal 1: Travel expense claim check support system - Automatically checks the validity of the requested route and amount - Detecting unusual routes and amounts - Automatic generation of check result reports Proposal 2: Labor Compliance Checker - Automatically detect legal violation risks in attendance data - Detecting signs of violation of the 36 Agreement - Automatically generate monthly compliance reports Proposal 3: Event-related document creation support - Proposal template generation - Drafting emails for event information - Support for designing questionnaire forms Proposal 4: Internal FAQ Automated Response System - Automated responses to common application questions - Presents answer candidates based on past answer history - Sorting and prioritizing new inquiries Proposal 5: Event Management Efficiency Assistant - Automatically create a management table for ordering supplies and lunch boxes - Proposal for optimizing order quantity according to the number of participants - Automatic generation of time schedule tables Proposal 6: Attendance Abnormality Detection System - Automatic detection of missed time stamps and input errors - Detecting abnormalities in work patterns - Confirmation priority determination and notification

[0044] The objective information is transmitted from the user terminal communication unit 30 to the user terminal 12, and multiple suggestions such as those described above are displayed on the user terminal 12. The user can select one or more of these suggestions and transmit them to the information processing system 10. The information processing system 10 can cause the generation AI system 15 to generate more specific instructions for utilizing the generation AI system for the suggestions selected by the user and transmit them to the user terminal 12.

[0045] Regarding the target information, the target information compiled in a certain format may be output from the information output unit 33. In this case, the information output unit 33 causes the generation AI system 14 to generate target information that summarizes the context information in a certain format via the generation AI system communication unit 31, and receives and outputs this target information. Also, in this case, in S6, the information processing system 10 may receive from the generation AI system 15 information indicating that the target information can be output, rather than the target information itself. Based on this information, the information processing system 10 determines in S7 that the target information can be output, and in S8, the information output unit 33 causes the generation AI system 15 to generate and output the target information.

[0046] Next, a second embodiment of the information processing system 10 will be described. In this embodiment, the target information of the information processing system 10 is data referenced by a Retrieval-Augmented Generation (RAG) system (hereinafter referred to as RAG data). To create the RAG data, the information processing system 10 confirms the target and information on which the RAG data is based from the user, and asks additional questions based on the answers to deepen the context information, thereby increasing the accuracy of the target information.

[0047] The configuration of the information processing system 10 is the same as that of the first embodiment shown in FIGS. 1 and 2. The operational flow is also the same as that of the first embodiment shown in FIG. 3. First, basic questions are sent from the user terminal communication unit 30 to the user terminal 12 (S1). The basic questions in this embodiment are questions for (1) confirming the purpose and (2) confirming the prescribed documents and basic information. In the purpose confirmation, questions are asked to confirm the main use of the RAG data and the users of the RAG system (e.g., internal or external customers). In the confirmation of the prescribed documents and basic information, questions are asked for the purpose of inputting data to create RAG data, such as uploading document files such as internal documents or manuals, entering a website URL, or having the user directly enter information in text.

[0048] Basic questions are sent to the user terminal 12, and the user inputs information (S2). When the user terminal communication unit 30 receives this information (S3), the prompt generation unit 32 generates an instruction prompt including a question generation instruction and a target information generation instruction (S4). At this time, the information analysis unit 34 analyzes and acquires information about the file sent from the user terminal 12. The information analysis unit 34 also acquires information by web scraping about the website URL sent from the user terminal 12. Using the basic questions, their answers, and the information acquired by the information analysis unit 34 as context information, the prompt generation unit 32 generates an instruction to create additional questions in the question generation instruction.

[0049] The purpose of the information to be generated, as stated in the instruction prompt, is to "gather comprehensive information through a series of questions based on the initial information provided by the user." The instruction prompt includes instructions to analyze the objectives and basic information that are the answers to the basic questions and to create additional questions to gather more comprehensive information about potential data sources.

[0050] The question generation instructions in the instruction prompt include the following steps for generating follow-up questions: (1) analyze the initial information and subsequent user responses, (2) create up to three follow-up questions, (3) present three example answers for each question, and (4) wait for the user's response before generating the next question.The question generation instructions also include points to keep in mind when generating follow-up questions, such as focusing on collecting and organizing data rather than implementing the RAG system, deepening understanding of existing information, and making questions specific and easy to answer.

[0051] The question generation prompts include instructions for the generative AI system15 to consider (1) summarizing key points, (2) identifying key themes and topics, (3) categorizing and evaluating potential data sources, (4) gaps and ambiguities in the information, (5) new perspectives to explore, (6) rationale for new questions, (7) data organization considerations, and (8) the suitability of text analytics and topic modeling. The prompts also include instructions to continue the process of analysis, question generation, and waiting for user responses until comprehensive information about potential data sources for the RAG system is gathered.

[0052] The instruction prompt includes a destination information generation instruction that generates destination information based on the acquired context information and outputs the destination information if the accuracy of the context information is above a certain level. In this embodiment, the destination information is output as text data that can be referenced by the RAG system. When a user inputs a question, the RAG system extracts the intent and keywords of the question through text analysis, and identifies to which topic in the topic classification master information the question belongs based on the extracted keywords and context information. Furthermore, the RAG system searches and acquires data sources related to the identified topic, and generates an answer based on the acquired information to respond to the user. In this embodiment, the destination information includes the topic classification master information and data sources used by the RAG system. Whether the accuracy of the context information is above a certain level can be determined by the generation AI system 15 based on whether the number or volume of items in the topic classification master information or data sources, which are the destination information, is above a certain level, or whether the number of additional questions generated has reached an upper limit.

[0053] The operations of the information processing system 10 after the prompt generating unit 32 generates the instruction prompt (S5 to S7, S9 to S11 in FIG. 3) are the same as those in the first embodiment, and therefore will not be described again.

[0054] In the second embodiment, when outputting information in S8, the information output unit 33 causes the generation AI system 15 to generate objective information by organizing the context information into a specific format via the generation AI system communication unit 31. In this embodiment, the information ultimately generated is RAG data, so the context information is organized in the generation AI system 15 into a RAG data format that can be recognized by the RAG system, thereby generating objective information. To achieve this, the prompt generation unit 32 generates a prompt that includes instructions to format and classify the context information, consisting of questions and answers accumulated up to that point, into content suitable for text analysis and topic modeling, and to construct a data source for use in the RAG system. The objective information generated by the generation AI system 15 is output from the information output unit 33.

[0055] Alternatively, the program used in the information processing system 10 of this embodiment can be provided independently. In this case, the program executes a user terminal communication process for sending and receiving information to and from the user terminal 12 used by the user, a generation AI system communication process for sending and receiving information to and from the generation AI system 15, a prompt generation process for generating an instruction prompt for the generation AI system 15, and an information output process for outputting target information in a certain format. The user terminal communication process generates an instruction prompt including a question generation instruction for causing the generation AI system 15 to generate an additional question based on context information consisting of pre-stored basic questions and answers corresponding to the basic questions. The prompt generation process generates an instruction prompt using the basic questions and answers corresponding to the basic questions as context information. The generation AI system communication process sends the instruction prompt to the generation AI system 15 and receives the generated additional question. The user terminal communication process sends the additional question to the user terminal 12 and receives the answer corresponding to the additional question. In the prompt generation process, an instruction prompt is generated that includes a question generation instruction that causes the generation AI system 15 to generate additional questions based on context information consisting of questions sent to the user terminal 12 and answers received from the user terminal 12, and a target information generation instruction that causes the generation AI system 15 to generate target information based on the acquired context information and enable output of the target information if the accuracy of the context information is above a certain level.In the information output process, if the information received from the generation AI system 15 in the generation AI system communication process includes target information or information indicating that the target information can be output, the target information is output.

[0056] Although the embodiments of the present invention have been described above, the application of the present invention is not limited to these embodiments and can be applied in various ways within the scope of its technical concept. In this embodiment, the target information is the method for utilizing the generation AI system and RAG data, but questions may be set and generated using other information as target information. For example, the resume required by a job seeker may be used as target information. Furthermore, other information may also be used as target information. [Explanation of symbols]

[0057] 10 Information Processing Systems 12 User terminals 13 Display means 15 Generative AI System 16. Communication Networks 20 Control Means 21 Memory means 22 Means of communication 30 User terminal communication unit 31 Generative AI System Communications Department 32 Prompt Generation Unit 33 Information output section 34 Information Analysis Department 40 Question display area 41 Answer input area 42 Upload Area 43 Button 50 Question display area 51 Choices 53 Button

Claims

1. An information processing system that acquires context information through a question to a user and outputs target information that can be created based on the context information and is information that meets a certain purpose, comprising control means and storage means; The storage means stores basic questions in advance, The control means a user terminal communication unit for transmitting and receiving information to and from a user terminal used by the user; A generation AI system communication unit that transmits and receives information to and from the generation AI system; a prompt generation unit that generates an instruction prompt for the generation AI system; an information output unit that outputs the target information in a certain format, the user terminal communication unit transmits the basic questions to the user terminal and receives answers corresponding to the basic questions; The prompt generation unit generates a first instruction prompt including a first question generation instruction for causing the generation AI system to generate an additional question based on the context information consisting of the basic question and the answer corresponding to the basic question; The generating AI system communication unit receives the additional questions generated by transmitting the first instruction prompt to the generating AI system; the user terminal communication unit transmits the additional questions to the user terminal and receives answers corresponding to the additional questions; the prompt generation unit generates a second instruction prompt including a second question generation instruction to cause the generation AI system to generate additional questions based on the context information consisting of the basic questions and the additional questions sent to the user terminal and the answers received from the user terminal, and a target information generation instruction to generate target information based on the acquired context information and enable output of the target information if the accuracy of the context information is equal to or higher than a certain level; the generation AI system communication unit receives the additional question or the target information generated by sending the second instruction prompt to the generation AI system, or information indicating that the target information is now available for output; The information output unit is an information processing system that outputs the target information if the information from the generation AI system received by the generation AI system communication unit includes the target information or information that the target information is now available for output.

2. The information processing system of claim 1, wherein the prompt generation unit includes in the first question generation instruction and the second question generation instruction an instruction to the generation AI system to generate multiple options that serve as answers to the additional question or the further additional question.

3. The information processing system of claim 1, wherein the prompt generation unit includes, in the target information generation instruction, an instruction to determine that the accuracy of the context information is above a certain level and enable the target information to be output when the number of items contained in the target information to be generated is above a certain level.

4. The information processing system of claim 3, wherein the prompt generation unit includes, in the target information generation instruction, an instruction to determine that the accuracy of the context information is above a certain level regardless of the number of items included in the target information, and to enable the target information to be output, when the number of additional questions already generated is above a certain level.

5. The information processing system according to any one of claims 1 to 4, wherein the objective information is information that suggests to the user how to utilize the generation AI system.

6. 5. The information processing system according to claim 1, wherein the target information is Retrieval-Augmented Generation (RAG) data referenced by a RAG system.

7. A program that acquires context information through questions to a user and creates purpose information that can be created based on the context information and is information that meets a certain purpose, On the computer, A user terminal communication process for transmitting and receiving information to and from a user terminal used by the user; A generation AI system communication process for transmitting and receiving information to and from the generation AI system; a prompt generation process for generating instruction prompts for the generation AI system; an information output process for outputting the target information in a predetermined format; In the user terminal communication process, a first instruction prompt is generated, the first instruction prompt including a first question generation instruction for causing the generation AI system to generate additional questions based on the context information consisting of pre-stored basic questions and answers corresponding to the basic questions; In the prompt generation process, the first instruction prompt is generated using the basic questions and the answers corresponding to the basic questions as the context information; In the generating AI system communication process, the first instruction prompt is sent to the generating AI system to receive the generated additional questions; In the user terminal communication process, the additional questions are sent to the user terminal and answers corresponding to the additional questions are received; In the prompt generation process, a second instruction prompt is generated, which includes a second question generation instruction for causing the generation AI system to generate additional questions based on the context information consisting of the basic questions and the additional questions sent to the user terminal and the answers received from the user terminal, and a target information generation instruction for causing the system to generate target information based on the acquired context information and enabling the system to output the target information if the accuracy of the context information is equal to or higher than a certain level; In the generation AI system communication process, the second instruction prompt is sent to the generation AI system, and the generated additional question or the target information or information indicating that the target information is ready to be output is received; In the information output processing, if the information received from the generation AI system in the generation AI system communication processing includes the target information or information indicating that the target information can be output, the program outputs the target information.

Images