Programs, information processing devices, methods, and systems

Abstract

According to this disclosure, it is possible to provide a chatbot system that enables users to seek advice on their problems. [Solution] A program that runs on a computer. This program has a processor that has a basic prompt definition that includes comprehensive instructions to control the flow of the entire dialogue. The program inputs trigger information into the generating AI system to start the dialogue, obtains the first comment from the generating AI system to elicit personal information and presents it to the user, inputs queries from the user for information gathering into the generating AI system, obtains a response from the generating AI system to extract information from a predetermined perspective and presents it to the user, when the generating AI system determines that predetermined termination requirements have been met, obtains a dialogue termination confirmation comment from the generating AI system and presents it to the user, and at closing, inputs the user's dialogue termination acceptance information into the generating AI system and obtains user evaluation information from the generating AI system.

Description

Technical Field

[0001] The present disclosure relates to a program, an information processing apparatus, a method, and a system.

Background Art

[0002] In recent years, companies and the like are required to understand the concerns and conditions of employees and smooth communication. For such purposes, the use of chatbot technology has also been considered.

[0003] In Patent Document 1, a technique for improving practical communication ability through role-play is described through a chatbot system that conducts conversations using artificial intelligence including a large language model. The system described in Patent Document 1 analyzes a user's utterance, assigns a score to a predetermined evaluation item, and presents feedback to the user together with the evaluation result.

Prior Art Documents

Patent Documents

[0004]

Patent Document 1

Summary of the Invention

Problems to be Solved by the Invention

[0005] However, the chatbot system described in Patent Document 1 trains communication ability in a specific situation. However, in Patent Document 1, accepting consultations about users' concerns is not described. In a given group, there may be members who have concerns, and there is a demand for a chatbot system that can accept consultations about concerns. [[ID=!]]

[0006] An object of the present disclosure is to provide a chatbot system that can accept consultations about concerns. [Means for solving the problem]

[0007] To solve the above problems, a program according to one aspect of the present disclosure is a program for operating a computer comprising a processor and memory, the program causing the processor to perform the following steps: input trigger information that triggers the start of a dialogue to a generating AI system in order to start the dialogue initiation phase with a user; obtain from the generating AI system an initial comment adjusted to elicit personal information from the user based on a basic prompt definition including comprehensive instructions for dialogue control and the trigger information, and present it to the user; input a query input by the user to the generating AI system in the dialogue information gathering phase; obtain from the generating AI system a response to elicit information from the user regarding a predetermined viewpoint based on the basic prompt definition and the query, and present it to the user; when the generating AI system determines that a predetermined termination requirement has been met based on the basic prompt definition and the content of the dialogue, obtain a comment from the generating AI system prompting the user to confirm that they wish to terminate the dialogue, and present it to the user; and in the dialogue closing phase, input information that the user has agreed to terminate the dialogue to the generating AI system; obtain user evaluation information from the generating AI system based on the basic prompt definition and the content of the dialogue, and present the evaluation information to the user. [Effects of the Invention]

[0008] According to this disclosure, it is possible to provide a chatbot system that enables users to seek advice on their problems. [Brief explanation of the drawing]

[0009] [Figure 1] This is a block diagram showing the overall configuration of System 1. [Figure 2] This is a block diagram showing an example of the functional configuration of the terminal device 10. [Figure 3] This block shows a functional configuration example for server 20. [Figure 4]This diagram shows the data structure of a table. [Figure 5] This diagram shows the data structure of a table. [Figure 6] This figure shows an example of the processing flow in System 1. [Figure 7] This figure shows an example of the screen in this disclosure. [Figure 8] This figure shows an example of the screen in this disclosure. [Figure 9] This figure shows an example of the screen in this disclosure. [Figure 10] This block diagram shows a functional configuration example of Server 20A. [Figure 11] This diagram shows the data structure of a table. [Figure 12] This diagram shows the data structure of a table. [Figure 13] This diagram shows the data structure of a table. [Figure 14] This figure shows an example of the processing flow in a usage example. [Figure 15] This figure shows an example of the screen in this disclosure. [Figure 16] This is a block diagram showing the basic hardware configuration of Computer 90. [Modes for carrying out the invention]

[0010] The embodiments of this disclosure will be described below with reference to the drawings. In all the drawings illustrating the embodiments, common components are denoted by the same reference numerals, and repeated explanations are omitted. The following embodiments are not intended to unduly limit the content of this disclosure as described in the claims. Not all components shown in the embodiments are necessarily essential components of this disclosure. Also, each drawing is a schematic diagram and is not necessarily a strict illustration.

[0011] Also, in the following description, a "processor" refers to one or more processors. At least one processor is typically a microprocessor such as a CPU (Central Processing Unit), but may also be other types of processors such as a GPU (Graphics Processing Unit). At least one processor may be single-core or multi-core.

[0012] Also, at least one processor may be a processor in a broad sense, such as a hardware circuit (e.g., FPGA (Field-Programmable Gate Array) or ASIC (Application Specific Integrated Circuit)) that performs part or all of the processing.

[0013] Also, in the following description, an expression such as "xxx table" may be used to describe information from which an output is obtained for an input. This information may be data of any structure or a learning model such as a neural network that generates an output for an input. Therefore, "xxx table" can be referred to as "xxx information".

[0014] Also, in the following description, the configuration of each table is an example. One table may be divided into two or more tables, or all or part of two or more tables may be one table.

[0015] Also, in the following description, when the "program" is used as the subject to describe processing, the program is executed by a processor to perform the defined processing while appropriately using a storage unit and / or an interface unit, etc. Therefore, the subject of the processing may be a processor (or a device such as a controller having that processor).

[0016] The program may be installed on a device such as a computer, or it may reside on a program distribution server or a computer-readable (e.g., non-temporary) recording medium. Furthermore, in the following description, two or more programs may be implemented as a single program, or one program may be implemented as two or more programs.

[0017] Furthermore, in the following explanation, identification numbers are used as identification information for various objects, but other types of identification information (for example, identifiers that include letters or symbols) may also be used.

[0018] Furthermore, in the following explanations, when describing similar elements without distinction, a reference code (or a common code among reference codes) may be used, and when describing similar elements with distinction, the element's identification number (or reference code) may be used.

[0019] Furthermore, in the following explanation, only control lines and information lines deemed necessary for the explanation are shown, and not all control lines and information lines in the product are necessarily shown. All components may be interconnected.

[0020] Each information processing device consists of a computer equipped with an arithmetic unit and a memory device. The basic hardware configuration of the computer and the basic functional configuration of the computer realized by said hardware configuration will be described later. For each of the terminal device 10 and the server 20, explanations that overlap with the basic hardware configuration and basic functional configuration of the computer described later will be omitted.

[0021] [System 1] <1. Overview> System 1 utilizes generative AI technology to provide an AI mentor service that responds to employee concerns and offers evaluations or advice. In this service, employees interact with an anthropomorphic conversation partner (AI mentor) driven by generative AI. The generative AI internally maintains basic prompt definitions, including the role of the AI ​​mentor and the logic or rules for each phase of the conversation (start, information gathering, closing), and autonomously controls the flow of the conversation and the generation of response content based on these. Server 20 primarily relays user input to the generative AI and presents the user with responses from the generative AI (comments, questions, evaluations, advice, etc.). The conversation progresses through a "start phase" where the AI ​​mentor communicates the purpose and creates a comfortable atmosphere, an "information gathering phase" that delves deeper into the employee's situation or feelings from a specific perspective, and a "closing phase" where the generative AI generates and presents evaluations and advice based on the conversation content, concluding the conversation. The generated evaluations can also be linked with the talent management system 50. Large-scale language models (LLMs) can be used for the generative AI.

[0022] <2. Overall System Configuration> Figure 1 is a block diagram showing an example of the overall configuration of System 1. This disclosure explains the system using an example in which a system for using the AI ​​mentor service is introduced within a corporate organization, employees belonging to that organization use the AI ​​mentor service as users, and the results are utilized in organizational management. The AI ​​mentor service may have a function to communicate in a friendly tone that evokes a specific region or character, or a casual conversational style that includes emojis, in order to make it easier to elicit honest opinions from employees. In addition, the AI ​​mentor service may combine initial information gathering in the form of questionnaires with detailed information gathering through dialogue based on the questionnaire results in order to efficiently and deeply elicit employees' honest opinions and specific challenges.

[0023] As shown in Figure 1, this embodiment includes a system 1 comprising, for example, a terminal device 10, a server 20, a generation AI system 30, and an internal company information database 40. System 1 is communicated via a network 80 with a talent management system 50, which is provided as a separate system.

[0024] Figure 1 shows an example where System 1 includes one terminal device 10, for the sake of illustration simplification. However, in reality, System 1 may include multiple terminal devices 10 for use by multiple users.

[0025] In Figure 1, System 1 is shown as an example that includes one Server 20, but for example, a collection of multiple devices may be considered as one Server 20. The way in which the multiple functions required to realize Server 20 are distributed to one or more hardware can be appropriately determined according to the processing capacity of each hardware and / or the specifications required for Server 20.

[0026] Figure 1 shows an example where System 1 includes one Generative AI System 30, but System 1 may include two or more Generative AI Systems 30. Also, Figure 1 shows an example where the Generative AI System 30 is independent of Server 20, but Server 20 may include the functions of the Generative AI System 30. In other words, Server 20 may store the LLM included in the Generative AI System 30.

[0027] Figure 1 shows an example where the internal information database 40 is independent of the server 20, but the server 20 may include the functions of the internal information database 40. In other words, the internal information database 40 may be built on the internal storage of the server 20. Similarly, the talent management system 50 may cooperate as an existing system existing outside of system 1, or it may be integrated as part of the functions of the server 20.

[0028] Terminal device 10 is, for example, an information processing device operated by an employee. Terminal device 10 may be implemented as, for example, a mobile device such as a smartphone or tablet, or a stationary PC (Personal Computer), laptop PC, etc. Employees access the AI ​​mentor service via a dedicated application or web browser installed on terminal device 10 and interact with the AI ​​mentor.

[0029] The terminal device 10 comprises a communication interface 12, an input device 13, an output device 14, memory 15, storage 16, and a processor 19. The input device 13 is a device (e.g., a touch panel, touchpad, mouse or other pointing device, keyboard, etc.) for receiving input operations from the user (employee) (text input of problems or consultation content, selection of responses, specification of tone, etc.). The output device 14 is a device (display, speaker, etc.) for presenting information such as comments, questions, evaluations, advice, and summaries from the AI ​​mentor to the user employee.

[0030] Server 20 is, for example, an information processing device for managing and operating the AI ​​mentor service, and is implemented by a computer connected to network 80. Server 20 may also be, for example, an API (Application Programming Interface) server.

[0031] The server 20 receives requests from the terminal device 10 (for example, an instruction to start the AI ​​mentor service, or input from a user during an interaction) and controls the sending and receiving of necessary information with the generating AI system 30.

[0032] The storage unit 202 of the server 20 (implemented by memory 25 and storage 26) stores data and programs used to provide the AI ​​mentor service. The programs include application programs for providing the AI ​​mentor service. The storage unit 202 may also store the original or a part thereof of this basic prompt definition, and an evaluation table 2021 that stores the evaluation results generated by the user-specific dialogue session or the generating AI system 30. The server 20 may, as necessary, provide the generating AI system 30 with a part of the basic prompt definition or related context information as part of the prompt at the start of the dialogue or at each step of the dialogue.

[0033] As shown in Figure 1, the server 20 includes a communication IF 22, an I / O IF 23, memory 25, storage 26, and a processor 29. The I / O IF 23 functions as an interface for an input device that receives input operations from the administrator of the AI ​​mentor service, and an output device that outputs information to the administrator.

[0034] The generation AI system 30 is, for example, a cloud server that has an LLM. The number of LLMs included in the generation AI system 30 may be one or multiple.

[0035] LLM (Language Modeling) is a single-modal natural language model built by training on large amounts of text data, and is used for many NLG (Natural Language Generation) tasks, such as generating responses to specific questions, automatically generating sentences, and summarizing text. LLM is an example of a generative AI model. Examples of LLMs include the following: • OpenAI: GPT-4 Google: Gemini 1.5 Flash Anthropic:Claude 3.5 Sonnet

[0036] The generating AI system 30, for example, has an LLM (Limited Language Model) and autonomously controls the generation of the overall flow of the dialogue (management of the start phase, information gathering phase, closing phase, transition of perspectives, determination of termination requirements, etc.) and response content (comments to employees, questions, evaluations, advice, summaries, etc.) based on pre-configured basic prompt definitions. When the server 20 receives a trigger to start the dialogue, input from the user (query), and, if necessary, relevant contextual information (e.g., search results from the internal information database 40) as part of the prompt, the generating AI system 30 comprehensively interprets these and its internal basic prompt definitions, generates a response (e.g., data in JSON format) corresponding to the next step in the dialogue, and sends it to the server 20. The dialogue may be structured based on multiple main "perspectives" or "topics," and may be designed in a "hierarchical" structure that progressively delves into related sub-items within each topic. The number of basic questions for each topic may also be fixed.

[0037] The basic prompt definition includes information such as the role of the AI ​​generating system 30, the logic for the progression of each phase of the conversation, the perspectives and confirmation items for information gathering, conversation progression rules, termination requirements, evaluation criteria, response format (e.g., JSON), etc., which the AI ​​generating system 30 uses to autonomously control the conversation. The basic prompt definition is either pre-configured within the AI ​​generating system 30 or stored in the memory unit 202 of the server 20, and is provided to the AI ​​generating system 30 as part of the prompt at the start of the conversation or as needed.

[0038] The generating AI system 30 autonomously generates responses to the user (comments, questions, evaluations, advice, etc.) based on information such as user queries or dialogue content sent from the server 20 and basic prompt definitions that it internally references, and outputs them to the server 20. The basic prompt definitions include various instructions to enable smooth dialogue with employees and to collect and provide appropriate information, and their main components include role instructions, response generation instructions, reference information specifications, query information specifications, output format instructions, etc.

[0039] Role instructions include text that specifies the role (position) that the LLM will play when generating responses. In this embodiment, role instructions are defined, for example, within the basic prompt definition as "You are a friendly AI mentor. Strive to engage in conversations that deeply empathize with employees and provide a sense of security," specifying an appropriate persona for the AI ​​mentor. These role instructions are maintained throughout all phases of the conversation, but in the closing phase, the nuances of the role may be adjusted according to the purpose of the phase, such as "You are the one who summarizes the content of the conversation so far and provides empathetic feedback to the employee."

[0040] The response generation instructions include text that instructs the LLM on what kind of response to generate. In this embodiment, the response generation instructions vary greatly in content depending on the associated dialogue phase (e.g., introductory phase, information gathering phase, closing phase) or perspective (e.g., job satisfaction, interpersonal relationships, salary) in the basic prompt definition. For example, in the introductory phase, the instructions might include "Generate a comment that explains the purpose and flow of the dialogue and prompts the employee to fill out the initial questionnaire." In the information gathering phase, the instructions might include "Assert open-ended questions that are empathetic to the employee's feelings and elicit specific anecdotes or factors related to the specified perspective," or conversation flow rules (e.g., "Each response must end with a question"). In addition, if the information is insufficient, the instructions might include "Generate a comment that requests permission for additional questions," or if a specific answer cannot be obtained, "Generate a comment that presents options and asks for reasons." In the closing phase, instructions may include: "Based on the conversation so far, evaluate the employee's situation from various perspectives (e.g., a 6-point scale) and propose specific next actions (three pieces of advice)," or "Generate a comment expressing gratitude for their cooperation in the conversation and concluding the meeting."

[0041] Reference information specifications include instructions for using information extracted from the internal information database 40, or placeholders indicating where to insert such extracted information. The basic prompt definition may define a general format or instructions for using such external information. This is primarily used in the information gathering phase to provide more specific and company-specific information in response to specific employee concerns or questions, or to generate questions based on relevant information. For example, it might take the form of, "Please answer the employee's question by referring to the following company regulations: {extracted company regulations}".

[0042] Query information specifications include instructions for the LLM to recognize and use the content of input text (queries) from the user (employee), or placeholders indicating where to insert the input text string, when such input text (queries) is included. The basic prompt definition may define general guidelines on how to interpret user queries and use them in response generation. This is used in many aspects of the interaction, especially in the information gathering phase when incorporating employee responses into subsequent prompts. For example, it might take the form of, "Generate an empathetic response and follow-up questions for the following statement from the employee, '{employee input text}'."

[0043] The output format instructions include instructions regarding the output format (template), structure, style, or elements to be included in the response generated by the LLM. In this embodiment, the output format instructions are defined in detail in the "basic prompt definition," and for example, instruct the entire response from the generating AI system 30 to be in JSON format and to include specific key-value pairs. Examples of specific keys that may be included in this JSON-formatted response include "type" (conversation type), "emotionScore" (sentiment index), "progress" (progress rate), and "content" (conversation content text).

[0044] "Type" (conversation type) is an identifier that indicates the type of content of the response output by the generating AI system 30 within the dialogue, and can take values ​​such as "chat" (continuation of normal conversation), "evaluation" (response containing evaluation information), and "advice" (response containing advice). "EmotionScore" (emotion index) is a numerical representation of the user's emotional state estimated by the generating AI system 30 from the user's statements or the overall atmosphere of the dialogue, and can take a higher value, for example, if the user has strong positive emotions. "Progress" (progress rate) is a numerical value that indicates the degree to which the dialogue covers or achieves predetermined viewpoints or confirmation items, especially in the information gathering phase, and can take values ​​from 0% to 100%, for example. In the basic prompt definition, control rules may be set such as "if there are unconfirmed items, progress must always be 80% or less." "Content" (conversation content text) is the main text message presented to the user, and the actual lines spoken by the AI ​​mentor are stored as a string. In the basic prompt definition, the text in the content section may also be instructed to follow a specific style, such as "always include a question mark (?) to make it a question."

[0045] Furthermore, the text within the content may include style instructions in the introductory phase, such as "Respond in a casual tone with emojis, as if you were a friendly character," and instructions in the information gathering phase, such as "Present questions one at a time and concisely." In the closing phase, instructions may include requests for more structured output, such as "Clearly distinguish between evaluations and advice, and present next actions in a numbered list." Instructions regarding the tone specified by the employee, instructions regarding the generation of a summary of the conversation, and instructions regarding the content of specific advice may also be incorporated into the "basic prompt definition" as part of or related to these output format instructions.

[0046] The internal information database 40 is a database that stores internal information that the generating AI system 30 can refer to in order to generate more contextual or accurate responses based on internal information. For example, the server 20 extracts relevant information from the internal information database 40 via the information retrieval module 2035 in response to a request from the generating AI system 30, the context of the current dialogue, or a user query, and provides the extracted information to the generating AI system 30 as part of a prompt (utilizing RAG (Retrieval-Augmented Generation) technology). This database may store, for example, internal regulations, welfare information, career path examples, and past anonymized Q&A knowledge in a searchable format. However, the use of this internal information database 40 is not an essential configuration of this embodiment, and the basic problem-solving function of the AI ​​mentor service can be provided without this database.

[0047] The talent management system 50 is an existing system that centrally manages employee personnel information, skills, career plans, past performance evaluations, etc. The server 20 has the function of transmitting user evaluation information recorded in its memory unit to the talent management system 50 and storing it in association with other information. The server 20 may also have the function of notifying designated reporting destinations based on employee evaluation information when an employee's evaluation meets a predetermined standard value.

[0048] Each information processing device or system, such as the terminal device 10, server 20, generation AI system 30, internal company information database 40, and talent management system 50, may be composed of a computer 90 equipped with a processing unit and a storage device. The basic hardware configuration of the computer 90 and the basic functional configuration of the computer 90 realized by said basic hardware configuration will be described later. Note that explanations of each component that overlap with the basic hardware configuration of the computer 90 and the basic functional configuration of the computer will be omitted.

[0049] <3. Configuration of terminal equipment> Figure 2 is a block diagram showing an example of the functional configuration of the terminal device 10. As shown in Figure 2, the terminal device 10 comprises a communication unit 120, an input device 13, an output device 14, an optional voice processing unit 17, a microphone 171, a speaker 172, a camera 160, a location information sensor 150, a storage unit 180, and a control unit 190. Each block included in the terminal device 10 is electrically connected, for example, by a bus. In this embodiment, the terminal device 10 primarily provides an interface for employees, who are users, to use the AI ​​mentor service.

[0050] The communication unit 120 performs modulation and demodulation processing for the terminal device 10 to communicate with other devices. The communication unit 120 performs transmission processing on the signal generated by the control unit 190 and sends it to an external source (for example, the server 20). The communication unit 120 performs reception processing on the signal received from the external source and outputs it to the control unit 190. As a result, the content of the employee's consultation or various instructions are sent to the server 20, and the terminal device 10 receives the AI ​​mentor's response, evaluation, advice, etc. from the server 20.

[0051] The input device 13 is a device for a user (employee) operating the terminal device 10 to input instructions or information. The input device 13 can be implemented, for example, by a touch-sensitive device 131 on which instructions are input by touching the operating surface. If the terminal device 10 is a PC, the input device 13 may be implemented by a reader, keyboard, mouse, etc. The input device 13 converts the instructions input by the user (for example, text input of the content of the problem, request to end the dialogue, feedback on the advice given, specifying the tone of comments, etc.) into electrical signals and outputs the electrical signals to the control unit 190. The input device 13 may also include, for example, a receiving port that accepts electrical signals input from an external input device.

[0052] The output device 14 is a device for presenting information to the user (employee) operating the terminal device 10. The output device 14 is implemented by, for example, a display 141. The display 141 displays data corresponding to the control of the control unit 190 (for example, the initial comment from the AI ​​mentor, questions for information gathering, confirmation of the end of the conversation, employee evaluation, advice, summary of the conversation content, etc.). The display 141 is implemented by, for example, an LCD (Liquid Crystal Display) or an organic EL (Electro-Luminescence) display.

[0053] The voice processing unit 17 performs, for example, digital-to-analog conversion processing of the voice signal. The voice processing unit 17 converts the signal provided from the microphone 171 into a digital signal and provides the converted signal to the control unit 190. The voice processing unit 17 also provides the voice signal to the speaker 172. The voice processing unit 17 is implemented, for example, by a voice processing processor. The microphone 171 receives voice input and provides the voice signal corresponding to the voice input to the voice processing unit 17. The speaker 172 converts the voice signal provided from the voice processing unit 17 into voice and outputs the voice to the outside of the terminal device 10. In this embodiment, these voice-related components can be used when an employee inputs a problem or seeks advice by voice, or when an AI mentor provides a response by voice.

[0054] Camera 160 is a device that receives light using a photodetector and outputs it as a shooting signal. In the AI ​​mentor service of this embodiment, the camera function is not essential, but it can be used if an extended function such as video call-style interaction is envisioned.

[0055] The location information sensor 150 is a sensor that detects the position of the terminal device 10, and is, for example, a GPS (Global Positioning System) module. A GPS module is a receiving device used in a satellite positioning system. In a satellite positioning system, signals are received from at least three or four satellites, and the current position of the terminal device 10, which is equipped with a GPS module, is detected based on the received signals. The location information sensor 150 may also detect the current position of the terminal device 10 from the position of the wireless base station to which the terminal device 10 is connected. In the AI ​​mentor service of this embodiment, the location information sensor is not essential, but it can be used, for example, when providing support information according to the region or location.

[0056] The storage unit 180 is implemented by, for example, memory 15 and storage 16, and stores data and programs used by the terminal device 10. The storage unit 180 stores, for example, user information 181, application programs for using the AI ​​mentor service, or configuration information.

[0057] User information 181 includes, for example, information about the user (employee) using the terminal device 10. User information includes, for example, the user's name, age, address, date of birth, contact information, AI mentor service usage history (if stored locally), display settings (default settings for font size, tone of voice, etc.).

[0058] The control unit 190 is realized when the processor 19 reads a program stored in the memory unit 180 and executes instructions contained in the program. The control unit 190 controls the operation of the terminal device 10. By operating according to the program, the control unit 190 performs the functions of an operation reception unit 191, a transmission / reception unit 192, and a presentation control unit 193.

[0059] The operation reception unit 191 processes instructions or information input from the input device 13. Specifically, for example, the operation reception unit 191 can receive instructions input from a touch-sensitive device 131 or the like, or information related to the AI ​​mentor service, such as text regarding employee consultations, selections regarding the progress of the conversation, and specification of tone of voice.

[0060] Furthermore, the operation reception unit 191 receives voice commands input from the microphone 171. Specifically, for example, the operation reception unit 191 receives voice signals input from the microphone 171 and converted into digital signals by the voice processing unit 17. This allows employees to interact with the AI ​​mentor using voice.

[0061] The transmitting / receiving unit 192 performs processing to enable the terminal device 10 to send and receive data with an external device such as the server 20 in accordance with a communication protocol. Specifically, for example, the transmitting / receiving unit 192 sends the content of the consultation input by the user (employee), the dialogue response, and various instructions (such as specifying tone of voice and requesting termination) to the server 20. The transmitting / receiving unit 192 also receives information provided by the server 20 (such as comments, questions, evaluations, advice, and summaries from the AI ​​mentor).

[0062] The presentation control unit 193 controls the output device 14 to present information provided by the server 20 to the user (employee). Specifically, for example, the presentation control unit 193 displays various information related to the AI ​​mentor service, such as the AI ​​mentor's responses, questions, confirmation of the end of the conversation, employee evaluation, advice, and conversation summary, on the display 141. The presentation control unit 193 also outputs the information sent from the server 20 through the speaker 172.

[0063] <4. Server Configuration> Figure 3 is a block diagram showing an example of the functional configuration of the server 20 shown in Figure 1. As shown in Figure 3, the server 20 performs the functions of a communication unit 201, a storage unit 202, and a control unit 203.

[0064] The communications unit 201 performs processing to enable the server 20 to communicate with external devices, such as the terminal device 10, or the generation AI system 30, the company's internal information database 40, or the talent management system 50.

[0065] The memory unit 202 is implemented by memory 25 and storage 26, and stores data and programs used by the server 20 to provide the AI ​​mentor service. The programs include application programs for providing the AI ​​mentor service. The memory unit 202 also stores, for example, an evaluation table 2021.

[0066] The evaluation table 2021 stores information such as each employee's dialogue sessions, perspectives (job satisfaction, interpersonal relationships, salary, etc.), and corresponding evaluation texts and evaluation indicators.

[0067] The control unit 203 is realized when the processor 29 reads a program stored in the memory unit 202 and executes the instructions contained in the program. The control unit 203 controls the operation of the server 20. By operating according to the loaded program, the control unit 203 can perform the functions of a receive control module 2031, a transmit control module 2032, a service processing module 2033, an information presentation module 2034, and an information retrieval module 2035.

[0068] The receiving control module 2031 controls the process by which the server 20 receives signals from external devices according to a communication protocol. For example, it receives operational information such as text input (queries) from employees on the terminal device 10 and acceptance of ending a conversation. It also receives responses output from the generating AI system 30 (data such as output instructions to the chatbot, evaluations, advice, summaries, etc., in JSON format, for example) indirectly or directly via the service processing module 2033. The received information is passed to other related modules (mainly the service processing module 2033).

[0069] The transmission control module 2032 controls the process by which the server 20 transmits signals to external devices according to a communication protocol. For example, it performs processes such as sending user-input queries, dialogue start triggers, or necessary contextual information as part of a prompt to the generating AI system 30 based on instructions from the service processing module 2033; sending responses (questions or comments, etc.) from the generating AI system 30 to the terminal device 10 (chatbot) based on instructions from the information presentation module 2034; sending evaluation information to the talent management system 50; or controlling reporting to a predetermined reporting destination.

[0070] The service processing module 2033 serves as the primary communication interface with the generation AI system 30. Specifically, it formats user input (queries), dialogue initiation triggers, or other necessary dialogue text information received from the terminal device 10 via the receiving control module 2031 as part of the prompt, and inputs it to the generation AI system 30 via the transmission control module 2032. It also receives responses output from the generation AI system 30 (text for display to the chatbot, or structured data such as evaluations and advice, instructions regarding dialogue control, etc., returned in JSON format, for example) via the receiving control module 2031. The acquired response data is passed to the information presentation module 2034 for processing for output to the chatbot, or to the transmission control module 2032 for processing for recording or external linkage.

[0071] The information presentation module 2034 processes the output content of the generating AI system 30 (user-oriented questions, comments, evaluations, advice, summaries, etc.) received from the service processing module 2033 into a format suitable for the chatbot interface used by the employee user (such as the display screen of the terminal device 10), and then transmits and presents it to the terminal device 10 via the transmission control module 2032.

[0072] The information retrieval module 2035 is an optional function that can retrieve relevant information from the internal information database 40 and provide the information for the generating AI system 30 to use in its dialogue. When this function is used, the generating AI system 30 analyzes the context of the current dialogue or the query entered by the user based on its internal basic prompt definition, and if it determines that it is necessary to refer to internal information, it communicates a request to that effect or the information necessary for the search to the server 20 (e.g., the service processing module 2033). Based on this request from the generating AI system 30 or the communicated information, the service processing module 2033 instructs the information retrieval module 2035 to search the internal information database 40. The information retrieval module 2035 extracts relevant information according to the specified conditions (e.g., a specific keyword or topic) and provides it to the service processing module 2033. The service processing module 2033 includes the extracted information in the next prompt and sends it to the generating AI system 30. This enables the generating AI system 30 to generate a more contextual or accurate response based on internal information. The information retrieval module 2035 may retrieve information related to a query entered by a user from the internal information database 40 and provide it to the service processing module 2033 if the query meets certain requirements. These requirements include, for example, the query containing certain phrases. However, the use of the information retrieval module 2035 and the internal information database 40 is not essential for the basic problem-solving function of the AI ​​mentor service, and the main functions of the AI ​​mentor service can still be provided even without them.

[0073] <5. Data Structure> This section describes the data structures used in System 1. Note that the data structures described are examples only, and do not exclude data not mentioned. Furthermore, even data represented as the same table may be stored in physically separate memory areas. The main data structures used in the AI ​​mentor service in this embodiment are described as the internal information database 40 and the evaluation table 2021.

[0074] Figure 4 shows an example of the data structure of an internal company information database 40 that the AI ​​mentor may refer to when generating a response in this embodiment. The purpose of this internal company information database 40 is to store internal company information that the AI ​​mentor may refer to as needed when generating advice for employees. The internal company information database 40 shown in Figure 4 has a data structure that associates information item IDs, which identify individual information items, with information categories as keys, and reference destinations.

[0075] The internal information database 40 stores multiple information items for each information category, managed in units suitable for reference by the AI ​​mentor. For example, if a specific information category is "employee benefits," one of the information items included in this category might be "regulations regarding childcare leave," which could describe the outline of the system, eligible individuals, application procedures, etc. These information items are edited and updated as needed by the HR staff or system administrator.

[0076] The "Information Category" field is used to store unique identifying information (e.g., category name, category code) for grouping individual information items into specific categories (e.g., "Personnel Regulations," "Employee Benefits," "Work Procedures").

[0077] The "Information Item ID" field is a field that stores identification information (e.g., document name, unique ID) to uniquely identify individual information items.

[0078] The "Reference Information" field stores information such as the specific content (text data, etc.) of an information item identified by the "Information Item ID" field, or the location where it is stored (e.g., file path, URL, specific record in a database).

[0079] As described above, the internal information database 40 shown in Figure 4 stores, for example, information categories, along with identification information and reference information for individual information items belonging to those categories. This allows the server 20 to efficiently identify highly relevant internal information based on the content of the interaction with the employee and include it in the prompts sent to the generating AI system 30.

[0080] Figure 5 shows an example of the data structure of an evaluation table 2021 that may be stored in the memory unit 202 of the server 20. As shown in Figure 5, the evaluation table 2021 may be configured as a table having evaluation text and main columns for evaluation indicators, with, for example, dialogue ID, employee ID, and perspective as keys. This table stores evaluations of each aspect of the employee, such as job satisfaction, interpersonal relationships, and salary, generated through dialogue with the AI ​​mentor.

[0081] The "Dialogue ID" field stores an identifier that uniquely identifies a single dialogue session with the AI ​​mentor.

[0082] The "Employee ID" field is used to store an identifier that uniquely identifies the employee being evaluated.

[0083] The item "Perspective" is a string that directly indicates the perspective to be evaluated, or an item that stores a perspective ID to uniquely identify it.

[0084] The item "Evaluation Text" is an item that stores text information generated by the generating AI system 30 based on the content of the conversation, to describe the employee's situation or feelings, and the key points of the relevant conversation from a specific perspective. This may include a record of the conversation during the interview with the employee (e.g., transcribed text data). This can also function as a summary for the employee to reflect on the content of the conversation. For example, from the perspective of "job satisfaction," it may store content such as, "It appears that the employee is dissatisfied with the lack of opportunities to acquire new skills in their current work."

[0085] The "Evaluation Indicators" item is a field that stores indicators generated by the AI ​​generation system 30 based on the content of the dialogue, in order to quantitatively indicate the employee's situation from a specific perspective. Examples of indicators include star ratings, multi-level scores, satisfaction levels, etc.

[0086] <6. Operation> The operation of System 1 in this embodiment will now be described. Figure 6 is a flowchart showing an example of the processing flow resulting from the cooperation between Server 20 and the Generating AI System 30 in this embodiment. The control unit 203 of Server 20 can perform functions such as a receiving control module 2031, a transmitting control module 2032, a service processing module 2033, an information presentation module 2034, and an information retrieval module 2035. In this embodiment, the Generating AI System 30 internally holds or can access a set of basic prompt definitions (which include detailed instructions such as the role of an AI mentor, the progress logic for each phase of the conversation, the viewpoints and confirmation items for information gathering, conversation progress rules, termination requirements, evaluation criteria, and response format (e.g., JSON)), and has the ability to autonomously control the flow of the entire dialogue and the generation of response content based on these.

[0087] In step S11, the server 20 receives an instruction from the user, an employee, to start the AI ​​mentor service by operating the terminal device 10 (receiving an instruction from the user to start a conversation). This instruction is received by the receiving control module 2031 and relayed to the service processing module 2033.

[0088] In step S12, the service processing module 2033 inputs an initial prompt to the generating AI system 30, which includes trigger information (e.g., a user identifier or a dialogue start command) that triggers the start of the dialogue, in order to initiate the "start phase" of the dialogue. Based on this input and its internal basic prompt definitions (e.g., setting the role of the AI ​​mentor, how to proceed with the start phase, or instructions to start with a light question such as "How have you been lately?"), the generating AI system 30 autonomously generates the first comment in the dialogue (e.g., an introduction by the AI ​​mentor, an explanation of the purpose of the dialogue, or the first question) and sends it to the server 20. The service processing module 2033 receives the first comment output from the generating AI system 30 (e.g., response data in JSON format) via the receiving control module 2031.

[0089] In step S13, the information presentation module 2034 formats the first comment output from the generation AI system 30 into a format suitable for the chatbot, then sends it to the terminal device 10 via the transmission control module 2032 for the employee to see (presentation of the first comment).

[0090] In step S14, when an employee enters a response text (query) from the terminal device 10, the receiving control module 2031 passes the user's query to the service processing module 2033. The service processing module 2033 inputs the received user's query as a prompt to the generating AI system 30. At this time, if the generating AI system 30 determines that it is necessary to refer to internal company information, the service processing module 2033 can also instruct the information retrieval module 2035 to extract relevant information from the internal company information database 40 and provide the generated AI system 30 with the extracted information included in the next prompt.

[0091] The generating AI system 30 autonomously advances the information gathering phase of the dialogue based on the input user query and its internally held basic prompt definitions. This includes setting the current perspective (e.g., job satisfaction) according to the basic prompt definitions, generating questions or empathetic words to elicit information about the given perspective (e.g., asking about confirmation items related to job satisfaction such as "how you feel about your current work," following conversation progression rules such as "each response must end with a question"), and managing metadata such as dialogue progress or emotion score. In particular, a "one-way progress tracking system" may be used to manage the dialogue progress. This system is characterized by providing progress indicators that always advance or maintain progress without being set back by user responses or actions. For example, progress is clearly calculated based on milestones such as completion of a questionnaire or completion of questions on each topic, and it maintains robustness against user response modifications and state changes. The calculated progress may be displayed visually on the user interface (e.g., as a bar graph and percentage). This clearly shows the user the overall picture of the conversation and their current position, contributing to increased motivation to complete the conversation and fostering a sense of accomplishment. Furthermore, in the information gathering phase, if there are items that received low ratings in the initial questionnaire, for example, the generating AI system 30 will prioritize extracting specific episodes or factors related to those items through the conversation.

[0092] The generating AI system 30 sends the generated response to the server 20. The service processing module 2033 receives this response (e.g., in JSON format) via the receiving control module 2031. The response formatted by the information presentation module is presented to the terminal device 10 via the transmission control module 2032, and awaits further responses from the employee. Step S14 may be repeated until the generating AI system 30 has obtained confirmation from the user regarding all items related to the perspectives defined in the basic prompt definition, and has obtained at least one episode from the user for each perspective, or the user has requested to end the dialogue, thus fulfilling the dialogue termination requirements (information gathering phase). The information gathering phase is intended to clarify the details of specific concerns and negative factors that are difficult to see with a simple stage evaluation, and to clarify the relationship between the evaluation value and actual dissatisfaction and issues.

[0093] In step S15, the generating AI system 30 autonomously determines whether the information gathering phase has met the predetermined termination requirements based on its internal basic prompt definition and the content of the conversation up to that point. If it determines that the termination requirements have been met, the generating AI system 30 generates a comment prompting the employee to confirm that they wish to end the conversation (e.g., "Well, you've told us a lot, but shall we wrap things up now?") according to the basic prompt definition (procedure for transitioning to the closing phase), and sends it to the server 20. The service processing module 2033 acquires this confirmation comment via the receiving control module 2031. The information presentation module 2034 presents the formatted confirmation comment to the terminal device 10 via the transmission control module 2032 (presenting the termination confirmation comment to the user).

[0094] In step S16, information indicating that the employee has agreed to end the conversation via the terminal device 10 is passed to the service processing module 2033 via the receiving control module 2031, and this agreement information is then transmitted from the service processing module 2033 to the generating AI system 30. Upon receiving this agreement, the generating AI system 30 autonomously generates an evaluation of the employee's situation or concerns and advice on future actions based on its internal basic prompt definitions (including the method of analyzing the conversation content so far, evaluation perspectives, direction of advice, evaluation criteria (e.g., 6-point scale), number of suggested next actions (e.g., 3)), and sends these to the server 20. The generating AI system 30 analyzes the conversation content and automatically detects or estimates organizational constraints (e.g., workload, resource shortages) and personal constraints (e.g., skills, time, psychological barriers) that the employee may be facing. Taking these detected constraints into consideration, it generates more realistic and actionable advice. The service processing module 2033 receives the response, including the evaluation and advice, via the receiving control module 2031 (evaluation and advice acquisition). The evaluation information and advice provided may also be compiled and provided as a practical feedback sheet to help employees objectively reflect on their situation and take concrete action. One of the aims of the generated evaluation information is to visualize the essential factors of issues that could not be captured by conventional quantitative surveys alone, and insights that tended to be buried in unstructured qualitative data, thereby bridging the gap between data collection and utilization.

[0095] In step S17, the employee evaluation and advice acquired by the service processing module 2033 are recorded as employee evaluation information in the memory unit 202 (for example, stored in the evaluation table 2021). Furthermore, the evaluation and advice formatted by the information presentation module 2034 are transmitted to the terminal device 10 via the transmission control module 2032 and presented to the employee (presentation and recording of evaluation and advice). Direct feedback to the employee themselves (presentation of evaluation and advice) is intended to promote self-understanding and provide insights that tend to be lacking in conventional systems. In addition, the generation AI system 30 can autonomously generate a summary of the dialogue content at other predetermined timings of the dialogue (e.g., at the end of the closing phase) based on the basic prompt definition and present it to the employee via the server 20.

[0096] Furthermore, if the urgency level determined by the system (e.g., an urgency level determination engine provided in the information processing layer) based on employee evaluation information, particularly dialogue content, meets a predetermined threshold, the transmission control module 2032 reports the existence of the employee, or an overview of their evaluation, and the determined urgency level (e.g., "high," "medium," "low," etc.) to a designated recipient (supervisor, HR department, etc.). At this time, in response to a request from the designated recipient, the module may also perform a process to specifically present to the recipient the information that formed the basis of the evaluation or urgency level determination (e.g., specific statements or circumstantial evidence). The "predetermined threshold" and determination logic for determining whether the urgency level of an employee's situation is reportable may be determined by evidence-based urgency determination based on multiple criteria, such as those detailed below.

[0097] The specific mechanism for determining urgency involves first evaluating the frequency and type of specific negative keywords (e.g., phrases suggesting harassment, expressions indicating a strong intention to resign) from the user's dialogue, or whether the user's sentiment score, estimated by a generative AI system, exceeds a predetermined threshold and indicates a negative state. Next, it evaluates whether evaluation indicators (e.g., star rating, score) across multiple evaluation perspectives obtained in the information gathering phase are significantly low individually or in combination, or whether they show a deterioration of a certain level compared to the previous evaluation. Furthermore, the system attempts to detect not only this explicit information but also utterance patterns and situations from the entire context of the dialogue that suggest the employee may be experiencing serious problems (e.g., mental health issues, significant impairments to work performance). These multiple evaluation criteria (keywords, sentiment score, evaluation indicators, utterance patterns, etc.) are comprehensively analyzed and evaluated based on a predefined set of rules or using a machine learning model to determine the urgency of the employee's situation. Importantly, this determination is not merely a result; the specific utterances, related data, or circumstantial evidence that formed the basis of the determination are explicitly recorded and linked. This allows HR personnel and supervisors to grasp not only information indicating a high level of urgency, but also the background information (evidence) that led to that decision. This enables a more objective and rapid understanding of the situation and supports the decision-making process for appropriate initial responses.

[0098] <7. Screen example> Figures 7 to 9 illustrate examples of the screen of the display 141 of the terminal device 10 in this disclosure.

[0099] Figure 7 shows an example of a chat-style interface screen where an employee, as a user, engages in a conversation with an AI mentor using the generation AI system 30 regarding their concerns. This screen is displayed on the display 141 of the terminal device 10, allowing the employee to proceed with the conversation with the AI ​​mentor.

[0100] Areas 1411, 1413, 1415, 1417, and 1419 are areas that display comments or questions sent by the AI ​​mentor (generating AI system 30) to the employee. In the example in Figure 7, during the dialogue initiation phase and information gathering phase, the AI ​​mentor sequentially displays questions to elicit the employee's situation or concerns and specific episodes. For example, area 1411 displays the initiation question, "I'd like some hints on what we can do to make employees feel more comfortable at work. Are you ready for a conversation?", area 1413 displays "How have you been lately?", area 1415 displays "Have you had any difficulties at work recently?", area 1417 displays "Please tell me a specific episode where you didn't feel particularly fulfilled at work recently.", and area 1419 displays "That might be because there's a big gap between what you want to do and reality. What kind of work specifically do you want to do as an engineer?" These are questions designed to delve deeper into the employee's concerns and aspects of "fulfillment." These comments and questions were output by the generating AI system 30 based on user queries and basic prompt definitions.

[0101] Areas 1412, 1414, 1416, and 1418 are areas that display the responses entered and sent by employees in response to questions or comments from the AI ​​mentor. In the example in Figure 7, area 1412 displays "OK!", indicating agreement to start the conversation; area 1414 displays "I'm quite tired", indicating the current situation; area 1416 displays a specific concern, "I don't find my work fulfilling"; and area 1418 displays a specific reason or anecdote for not finding fulfillment, such as "What I want to do is more engineering-oriented work, not consulting, but I'm not able to do that at all." These employee input texts are sent to server 20 and become information for the AI ​​system 30 to generate the next response from the AI ​​mentor.

[0102] In addition to the basic chatbot screen shown in Figure 7, the interface screen of this embodiment may also include, for example, an area where the AI ​​mentor's tone of voice can be selected before the start of the conversation, or an area that displays a summary of the conversation content at the end of the closing phase of the conversation to encourage employee reflection.

[0103] Figure 8 shows an example of a screen that presents an evaluation based on the conversation with the AI ​​mentor and recommended next actions to the employee after the conversation has ended. This screen is displayed on the display 141 of the terminal device 10, allowing the employee to review the conversation and check the feedback from the AI.

[0104] At the top of the screen shown in Figure 8, introductory text indicating the purpose of the screen may be displayed, such as "Here's a summary of what we discussed today." Below this, multiple evaluation areas are provided. Area 1421 is an area that displays evaluations related to a predetermined perspective, "job satisfaction." Here, along with the perspective name "job satisfaction," an evaluation index calculated by the generating AI system 30 based on the content of the conversation, and evaluation text summarizing the employee's situation regarding that perspective, or the key points of the conversation, are displayed. In the example in Figure 8, the evaluation index is shown using a 5-star rating system.

[0105] Similarly, area 1422 is an area that displays an evaluation related to a predetermined perspective, "Human Relations," and displays the perspective name "Human Relations," evaluation indicators, and evaluation text. Area 1423 is an area that displays an evaluation related to a predetermined perspective, "Salary," and displays the perspective name "Salary," evaluation indicators, and evaluation text. These evaluation indicators and evaluation texts are generated by the generating AI system 30 by analyzing the content of the dialogue based on instructions from the server 20.

[0106] Area 1424 is the area that displays "Next Actions," which are specific action plans or advice recommended to employees. In the example in Figure 8, multiple actions are displayed in a numbered list format under the heading "Next Actions." These Next Actions are generated by the generating AI system 30 based on instructions from the server 20, taking into account the content of the dialogue and the evaluation results, and are tailored to each individual employee.

[0107] Figure 9 shows an example of a screen that displays information on employee satisfaction ratings and turnover probability for a specific employee. This screen is used by managers to understand the employee's situation and consider appropriate actions.

[0108] Area 1431 is a table that displays employee satisfaction ratings for each aspect, indicated as "Satisfaction." This table lists "Job Satisfaction," "Interpersonal Relationships," and "Salary" as aspects. For each aspect, the employee's evaluation score and the supervisor's evaluation score are displayed. In the example in Figure 9, the evaluation score is displayed as a number out of six levels, such as "6 / 6," "5 / 6," and "1 / 6." The employee's score is an evaluation index output by the AI ​​system 30 based on information collected through interactions with an AI mentor, for example, while the supervisor's score may display the supervisor's evaluation or target value, for example. Below each score, the change from the previous evaluation is displayed as "Previous Change," making it possible to understand the changes over time. Although not shown in the diagram, an interface may be provided to display the "Factors" that led to each satisfaction score.

[0109] Area 1432 is an area that displays the degree of likelihood of an employee leaving the company, indicated as "resignation probability." Area 1432 may display, for example, an indicator related to resignation estimated by Server 20 based on evaluation information with the employee stored in the memory unit. This resignation probability is calculated by comprehensively analyzing the content of conversations with the AI ​​mentor, satisfaction evaluations from various perspectives, etc., and is important information for managers to detect employee turnover risk early.

[0110] As shown in Figure 9, managers can objectively understand the current state and changes in employee satisfaction levels across multiple perspectives, such as job satisfaction, interpersonal relationships, and salary, as well as the likelihood of employee turnover, based on data. This information is aggregated in the talent management system 50 and used to support strategic decision-making in human resources, such as appropriate follow-up with employees, improving the quality of one-on-one interviews, and considering transfer or training plans. Furthermore, employees who meet certain criteria (e.g., significantly low satisfaction, high probability of turnover) may trigger notifications from the system to designated reporting destinations.

[0111] <8.Summary> As described above, in System 1, Server 20 first has the Generating AI System 30 generate initial comments to elicit personal information from the employee, who is the user, during the initial dialogue phase. Next, in the information gathering phase, based on the employee's responses, Server 20 has the Generating AI System 30 generate questions or responses to extract information on predetermined perspectives (job satisfaction, interpersonal relationships, salary, etc.). Then, after the dialogue meets the predetermined termination requirements and the employee's consent is obtained, in the closing phase, Server 20 has the Generating AI System 30 generate an evaluation of the employee and specific advice based on the content of the dialogue up to that point, and presents these to the employee. In this way, by combining a structured dialogue phase with perspective-based information gathering and the generation of individualized evaluations and advice, it becomes possible to address the employee's concerns more deeply and efficiently. This makes it possible to provide a chatbot system that enables employees to seek advice on their concerns. Furthermore, direct feedback to the employee (presentation of evaluations and advice) is intended to promote self-understanding and provide insights that tend to be lacking in conventional systems. Through interactions with the AI ​​mentor and the evaluations, advice, and summaries provided, employees can gain new insights into their own situation, emotions, and challenges from an objective perspective, which helps them to find concrete action plans.

[0112] [Examples of how evaluation information obtained from System 1 can be used] Next, we will explain examples of how evaluation information obtained through the AI ​​mentor service can be used. The generated interview support information provides an objective perspective based on AI analysis, addressing issues such as the lack of objectivity in prioritizing decisions that traditional HR personnel and supervisors should make. Furthermore, the interview support information for supervisors aims to strengthen collaboration between the generating AI system 30 and human interviewers, and to compensate for the lack of specific guidance for traditional follow-up interviews.

[0113] This section describes an example in which the server 20A instructs the generating AI system 30 to generate "first points" to be confirmed during an interview, or individually optimized interview support information (including how to start, how to explore solutions, how to encourage action, and how to close the interview), thereby generating interview support information to make the one-on-one interview conducted by the employee's supervisor more effective. The first points are several important points or themes that the supervisor should particularly confirm or explore during a one-on-one interview with a subordinate, output by the generating AI system 30 based on the employee evaluation information collected by the AI ​​mentor service, and include at least one of the following: identifying factors that give the employee a sense of fulfillment, confirming the employee's interest in or desire for new challenges, and reviewing the employee's current work content. Furthermore, this example describes how to generate interview support information consisting of items such as "how to start," "how to explore solutions," "how to encourage action," and "how to close the meeting," but the items generated as interview support information are not limited to these. In addition, it may be possible to suggest a realistic way to conduct the interview and prioritize topics, taking into account the constraints of the employee (e.g., lack of time, lack of specific skills, etc.) or the constraints of the supervisor (e.g., short interview time, etc.) extracted from the conversation with the AI ​​mentor.

[0114] The following is an overview of each item included in the interview support information. "Starting Method" (icebreaker) is advice for the introductory stage to help the supervisor start communication with the subordinate smoothly at the beginning of the interview. The generating AI system 30 takes into account the subordinate's personality traits (e.g., FFS diagnostic results), growth image, or information obtained from the conversation with the AI ​​mentor (e.g., specific emotions or situations expressed by the subordinate) and proposes effective ways to start the interview or examples of specific things to say. For example, it proposes questions that encourage understanding of the current situation, such as, "I heard that you haven't been feeling much fulfillment in your work lately, but when exactly do you feel that way?" or words that take the subordinate's characteristics into consideration, such as, "I feel like you're taking on too much because you have such a strong sense of responsibility," with the aim of creating an atmosphere where the subordinate can feel safe and comfortable talking.

[0115] "How to Find Solutions" is the main part of the interview, providing advice for the supervisor and subordinate to work together to find solutions regarding pre-identified perspectives (e.g., the subordinate's concerns, or key points to check as presented by the Generating AI System 30). The Generating AI System 30 considers the subordinate's personality or growth aspirations and provides specific examples of questions or approaches on how the supervisor should ask and how to explore solutions together with the subordinate. For example, collaborative suggestions such as, "Let's review your current work a little and think together about whether there are any parts that you can enjoy more," or questions that clarify the problem, such as, "What is the area you most want to improve in the current situation?" are suggested according to the subordinate's characteristics. In suggesting solutions, the system is designed to avoid falling into idealistic notions that do not consider organizational constraints or one-sided advice that relies solely on self-help, and instead presents more specific and actionable approaches.

[0116] "Encouraging Action" is advice given at the stage of translating the content discussed or solutions found in "Exploring Solutions" into concrete action plans that subordinates can implement, and encouraging their execution. The generating AI system 30 provides hints on how to translate the results of the discussion into concrete actions, points to consider when encouraging subordinates to take action, and even specific action suggestions tailored to the subordinate's growth image or characteristics (e.g., "Let's try to think of ways to incorporate a new perspective into our current work," a challenge within a reasonable scope).

[0117] The "Closing Method" is the final stage for effectively concluding the interview. The generating AI system 30, taking into account the subordinate's characteristics, growth vision, and the content discussed in the interview, presents examples of phrases to summarize the interview, reaffirm agreed-upon actions, or suggestions for the next follow-up. For example, depending on the subordinate's type, it may suggest phrases that respect the subordinate's growth vision, such as, "Let's try new things little by little, while valuing your own pace. There's no need to push yourself too hard," or phrases that encourage action, such as, "I think the ideas you shared today are interesting! Let's try them right away."

[0118] In this example, the same system configuration as System 1 shown in Figure 1 can be used. However, the information retrieval module 2035 in the control unit 203 can be configured arbitrarily. Server 20A has a program for executing this example in its storage unit 202, and the processor 29 executes it. However, as shown in Figure 10, in addition to the evaluation table 2021 that stores employee evaluation information obtained from the AI ​​mentor service, the storage unit 202A of Server 20A may also store data specific to this example, such as a thinking information table 2022, an interview support instruction table 2023, and an interview support information table 2024. The basic prompt definitions referenced by the AI ​​mentor service are either held internally in the generating AI system 30 or provided as appropriate from the storage unit 202A of Server 20A.

[0119] <Data structure in usage examples> Next, we will explain the main data structures used in this example.

[0120] Figure 11 shows an example of the data structure of a thought information table 2022 that can be stored in the memory unit 202A of server 20A or obtained from an external system such as a talent management system 50. This table stores information useful for personalizing the interview process, such as an employee's personality and growth image. As shown in Figure 11, the thought information table 2022 has a data structure that associates personality parameter information and growth image information using, for example, an employee ID as a key.

[0121] The "Employee ID" field stores an identifier that uniquely identifies the employee in question.

[0122] The "Personality Parameter Information" item stores the employee's personality traits. For example, it may store multiple parameter values ​​obtained based on specific methods such as the FFS (Five Factors & Stress) assessment, or text descriptions interpreting those values. Information representing an employee's personality can broadly include information obtained from diverse sources, such as the results of other standardized personality tests, descriptive feedback on personality based on multifaceted evaluations from supervisors and colleagues, or personality tendencies estimated from the employee's own self-analysis and daily behavioral records. The FFS (Five Factors & Stress) assessment is a theory and diagnostic method that analyzes an individual's thinking and behavioral characteristics or stress response based on "five factors" (generally cohesiveness, receptiveness, discriminativeness, diffuseness, and preservation). It is used to objectively grasp an individual's potential strengths or personality and for self-understanding, communication improvement, and skill development.

[0123] The "Growth Image Information" item stores information indicating the career growth image or values ​​that employees aspire to. For example, categories such as "desire rapid growth," "desire steady growth," and "desire to maintain the status quo," or free-form descriptions related to these, may be stored based on the results of a predetermined survey.

[0124] Figure 12 shows an example of the data structure of the interview support instruction table 2023 that can be stored in the memory unit 202A of server 20A in this application example. As shown in Figure 12, the interview support instruction table 2023 can be configured as a table having, for example, an instruction ID as the key, and columns for instruction type and instruction text (prompt template). Based on employee evaluation information obtained from the AI ​​mentor service and separately acquired information on the employee's thinking, this database defines the instruction content (components of the prompt) for the AI ​​system 30 to generate interview support information for supervisors (first perspective (perspectives to be confirmed in the interview) and how to proceed with the interview) from server 20A.

[0125] The item "Instruction Document ID" is an item that stores an identifier to uniquely identify instruction documents corresponding to each instruction task in the generation of supervisor interview support information (e.g., "Task to generate the first perspective (perspective to be confirmed in the interview)", "Task to generate the interview procedure").

[0126] The item "Type of Instruction" indicates the type of instruction, such as whether the instruction is intended to "generate the first perspective (perspective to be confirmed in the interview)" or "generate a plan for conducting the interview."

[0127] The item "Instruction Text (Prompt Template)" is an item that stores a template of the instruction text that is incorporated when the server 20A configures a prompt for the generation AI system 30. Specifically, the instruction text is string information that should be included in the prompt, and may include information for each part as described below, such as role instructions, response generation instructions, reference information specifications (inserting evaluation information obtained from the AI ​​mentor service or information about the employee's thinking, etc.), query information specifications (in this case, specific requests or situation settings from the supervisor, etc.), and output format instructions.

[0128] Role instructions include text that specifies the role (position) the LLM will take when generating responses. For example, role instructions include sentences that specify a persona suitable for generating interview support information, such as, "You are an experienced manager development consultant. You will deeply understand the situation and characteristics of your subordinates and generate specific advice to help their superiors conduct effective interviews."

[0129] The response generation instructions include text that instructs the LLM on what kind of response (interview support information) to generate. For example, if the "Instruction Type" is "Generate First Perspective," the response generation instructions might include instructions such as, "Based on the following employee evaluation information, please suggest three important perspectives that the supervisor should delve into in the next interview to help resolve the subordinate's concerns and support their growth, in the form of specific questions." Also, if the "Instruction Type" is "Generate Interview Procedure," the response generation instructions might include instructions such as, "Based on the presented 'First Perspective to Confirm' and the following 'Information on the Employee's Thinking,' please suggest a specific interview procedure tailored to the employee's personality traits and growth image, including how to start the interview, how to explore solutions for each perspective, how to encourage specific actions, and how to proceed until closing." Based on these instructions, server 20A assembles the appropriate prompt and sends it to the generating AI system 30.

[0130] The reference information specification includes instructions for the LLM to recognize and use employee evaluation information stored in the server 20A's memory unit 202, and employee personality traits, growth image, etc., obtained from the thinking information table 2022, or placeholders indicating where to insert such information, which are entered into prompts that the server 20A sends to the generation AI system 30. For example, specific data is incorporated into the prompt in the format of "Employee evaluation information summary: {evaluation information summary text}", "Employee personality traits (FFS): {FFS analysis results}", "Employee growth image: {growth image information}".

[0131] The query information specification includes instructions for the LLM to recognize and use parameters necessary for processing, such as a specific request from the supervisor that triggered the generation of this interview support information, the ID of the subordinate employee in question, or a placeholder indicating where to insert such information, which is entered at the prompt.

[0132] The query output format instructions include instructions regarding the output format, structure, style, or elements to be included in the response for the interview support information (first perspective and interview procedure) generated by the LLM. For example, the instructions may include: "Please present the first perspective as a numbered list," or "Please describe the interview procedure by dividing it into the following sections: 'How to start the interview,' 'How to explore solutions,' 'Specific actions,' and 'Closing,' each including specific recommended phrases."

[0133] Figure 13 shows an example of the data structure of the interview support information table 2024, which may be stored in the memory unit 202A of the server 20A. This table stores the content of the interview support information for supervisors output by the generating AI system 30 based on the information in the thinking information table 2022. The interview support information table 2024 stores, for example, the generated first perspective (perspectives to be confirmed in the interview) and specific interview support information, associated with the target employee ID.

[0134] The "Employee ID" field stores an identifier that uniquely identifies the subordinate employee being interviewed.

[0135] The item "List of First Perspectives" stores a list of multiple first perspectives to be confirmed during the interview, or text information for each perspective, output by the generating AI system 30. For example, multiple perspectives such as "Identifying factors that contribute to job satisfaction" and "Confirming interest in and desire for new challenges" may be stored.

[0136] The item "Interview Support Information" stores structured information or text information about the specific steps of conducting an interview, output by the generating AI system 30. This includes specific instructions or recommended communication methods for at least "How to start the interview," "Steps to explore solutions," "Steps to encourage specific actions," and "Closing steps." This information may be stored, for example, as detailed instructions for each step in JSON format, or as formatted text. <Operation in the example of use>

[0137] Next, we will explain the processing flow related to the generation of interview support information for supervisors, which is executed by server 20A. Figure 14 is a flowchart showing an example of the processing flow in this use case. This process can be initiated, for example, when an employee's supervisor requests support from the system to conduct interviews with subordinates effectively, or at a predetermined time (e.g., before a regular interview). The control unit 203 of server 20A, centered around the service processing module 2033, works in cooperation with modules such as the receive control module 2031, the transmit control module 2032, and the information presentation module 2034 to configure prompts for outputting interview support information in this use case, and executes the processing of this flow by sending and receiving information with the generation AI system 30.

[0138] In step S21, the service processing module 2033 acquires evaluation information about the subordinate employee being interviewed. This evaluation information concerns the user's problems, obtained through dialogue with the AI ​​mentor, and is stored, for example, in the storage unit 202A of the server 20A as an evaluation table 2021. Specifically, it may include evaluation text and indicators related to aspects such as job satisfaction, interpersonal relationships, and salary, or advice (next actions) presented to the employee, generated by the AI ​​mentor. The proposed advice may be structured to include a multi-layered solution approach, such as things the employee can do individually (individual level), things that can be improved in relationships with others (interpersonal level), and suggestions that lead to team or organizational-wide efforts as needed (organizational level), rather than being biased towards a single perspective. Furthermore, the proposed solutions and next actions are not limited to one method, and multiple feasible alternatives may be presented depending on the situation. Furthermore, to facilitate the implementation of proposed actions, they may be presented as a phased implementation plan broken down into specific steps (for example, what can be done in the short term, what to aim for in the medium to long term, etc.).

[0139] In step S22, the service processing module 2033 generates a first prompt that includes the employee evaluation information acquired in step S21 and instructions that pre-define in the interview support instruction table 2023 of the memory unit 202A, instructing the output of multiple first perspectives. The service processing module 2033 then inputs this first prompt to the generation AI system 30 via the transmission control module 2032, causing the generation AI system 30 to output multiple first perspectives. The output multiple first perspectives are received via the reception control module 2031 and held by the service processing module 2033 for use in subsequent processing.

[0140] In step S23, the service processing module 2033 acquires information about the employee's thinking. This information includes at least information representing the employee's personality (e.g., personal characteristics information such as FFS diagnostic results) and information representing the employee's growth image (e.g., survey results regarding values ​​for growth). This information about thinking can be acquired from the talent management system 50 or from a thinking information table 2022, etc., which is separately stored in the memory unit 202A of the server 20A.

[0141] In step S24, the service processing module 2033 generates a second prompt that includes multiple first perspectives output in step S22, information about the employee's thoughts acquired in step S23, and instructions that instruct the output of interview support information, which are predefined in the interview support instruction table 2023 of the memory unit 202A. The service processing module 2033 inputs this second prompt to the generation AI system 30 via the transmission control module 2032 and obtains interview support information from the generation AI system 30. The outputted interview support information is received via the reception control module 2031 and held by the service processing module 2033 for use in subsequent processing.

[0142] Interview support information includes, at a minimum, advice on how to start the interview (icebreaker), how to explore solutions, how to encourage action, and closing. This information can be more finely and individually optimized depending on the combination of the employee's personality traits (for example, traits such as cohesiveness (A trait), receptiveness (B trait), discriminativeness (C trait), diffusiveness (D trait), and preservativeness (E trait) based on the FFS (Five Factors & Stress) assessment) and the growth image the employee aspires to (for example, categories such as "reckless," "steady," "gradual," and "maintain the status quo"). This individual optimization is achieved by the generating AI system 30 comprehensively analyzing the information on the target employee's FFS traits and growth image, and generating the optimal communication scenario for each combination pattern. The generating AI system 30 may also select and prioritize the first perspective for outputting interview support information based on information representing the employee's personality or information representing their growth image (values).

[0143] In step S25, the service processing module 2033 stores the interview support information acquired in step S24 in the talent management system 50, associated with the target employee and their supervisor, either in the storage unit 202A of the server 20A (for example, as an interview support information table 2024) or via the transmission control module 2032, in order to present it to the user's supervisor. Subsequently, the information presentation module 2034 formats the stored interview support information (or interview support information directly linked from the service processing module 2033) into a format suitable for the screen of the terminal device 10 used by the supervisor, and transmits and presents it to the terminal device 10 via the transmission control module 2032. This allows the supervisor to view the interview support information as needed during interviews with the target employee.

[0144] <Specific usage examples> The interview support information for supervisors generated by Server 20A can be used in a variety of real-world management scenarios. Specific examples are given below.

[0145] Case Study 1: Addressing Communication Challenges Between Busy Supervisors and Subordinates Suppose a new employee (subordinate) expresses in a conversation with an AI mentor that they are troubled because their supervisor is always busy and they cannot secure time for consultations, and one-on-one meetings are frequently canceled. Based on this information, Server 20A generates and presents the supervisor with the following consultation support information: As a suggestion for the points to be checked (first point), it suggests, for example, "the difficulty the subordinate feels in securing time for consultations," "methods that have been effective in short-term communication with the subordinate in the past," and "the specific types of guidance or feedback the subordinate is seeking." As a suggestion for how to proceed with the consultation, for example, considering the time constraints, it suggests, for example, "a consultation plan to achieve results in a short time." It also considers the subordinate's personality traits (for example, if the FFS assessment analyzes that the D trait (diffusiveness) is strong) and recommends an approach such as "discussing concrete action plans that can be implemented immediately."

[0146] Case Study 2: Addressing Increased Workload Due to Subordinates' Insufficient Skills Suppose a conversation between a mid-level employee (subordinate) and an AI mentor reveals that the employee is suffering from chronic overtime due to increased workload caused by a lack of skills resulting from the introduction of a new system. Server 20A generates and presents the supervisor with the following interview support information: As a suggestion for the points to be checked (first point), it suggests, for example, "learning barriers that the subordinate specifically feels," "learning methods that the subordinate has found effective in the past," and "the priorities of tasks that the employee perceives." As a suggestion for how to conduct the interview, it suggests, for example, "an interview structure to explore a balance between short-term workload reduction measures and long-term skill improvement measures." Furthermore, considering the subordinate's personality traits (for example, the E trait (preservation) in the FFS assessment) and growth image (for example, aiming for "steady growth"), it recommends a dialogue approach such as "together considering how to reliably acquire each skill."

[0147] Case Study 3: Supporting a New Manager with Subordinates Having Diverse Personality Traits Suppose a newly appointed manager is struggling to conduct effective one-on-one interviews with multiple subordinates, each possessing different personality traits. Based on evaluation information output from the AI ​​mentoring service for each subordinate, as well as separately acquired personality trait and value information (FFS traits and growth image, etc.), Server 20A generates personalized interview support information (interview procedures) for each subordinate, suggesting communication tailored to their FFS traits. For example, to begin an interview (icebreaker), for a subordinate with trait E, it would be suggested to start with a question to understand the current situation and identify areas for improvement, such as, "I heard you've been feeling less fulfilled in your work lately. Specifically, when do you feel that way?" To explore solutions (example questions), for a subordinate with trait D, it would be suggested to ask a question that would pique their interest in new challenges, such as, "What skills would you like to develop by participating in a new project?" As for suggesting specific actions, for a subordinate with trait E who has a "gradual growth image," it would suggest a new challenge within a reasonable scope, such as, "Let's think about ways to incorporate a slightly new perspective into your current work." As an example of a closing, for an employee with the E trait who has a "gradual growth image," it is suggested to conclude with words that respect the employee's growth image, such as, "Let's take things at your own pace and gradually try new things. There's no need to push yourself too hard." This will help even newly appointed managers to quickly implement effective communication tailored to the characteristics of each employee.

[0148] As these examples illustrate, by combining the internal information of employees obtained through the AI ​​mentor with information such as individual personality traits or values, supervisors can gain concrete clues for conducting personalized, optimized communication with each subordinate, contributing to higher-quality one-on-one meetings. This also allows for the acquisition of concerns from group members through a chatbot system, and the acquired information can be used for group activities.

[0149] <Example screen in usage examples> Figure 15 shows an example of a meeting support information screen presented to a supervisor during a meeting with a specific subordinate employee. This screen displays advice generated by the AI ​​system 30 generated by the server 20A, based on information obtained through dialogue with the AI ​​mentor and separately collected employee characteristic information.

[0150] The top of the screen may display basic information such as the name of the subordinate employee or the date and time of the interview. Area 1441 is an area that displays a summary of the worries or concerns extracted by the AI ​​mentor from the conversation with the employee. This allows the supervisor to understand the subordinate's current main concerns before the interview.

[0151] Area 1442 is an area that displays the results of the personality analysis of the target employee as "individual personality traits" (for example, traits based on the FFS diagnosis: tendencies towards cohesiveness, receptiveness, discriminativeness, diffuseness, and preservation).

[0152] Area 1443 is a region that displays the type of career growth that the target employee aspires to, as an "image of individual growth" (for example, relentless, steady, gradual, maintaining the status quo, etc.).

[0153] Domain 1444 is a domain that presents important points (first perspective) that should be particularly checked during the interview, as generated by the AI ​​system 30 generated by Server 20A, in order to help resolve the subordinate's concerns. For example, perspectives such as "review of current work content," "degree of interest or desire for new challenges," and "specific factors that give a sense of fulfillment" may be listed.

[0154] Domain 1445 is an area that presents concrete communication plans as interview support information. This may include subsections such as how to start the interview (icebreaker), how to explore solutions, and closing. More specifically, for starting the interview (icebreaker), based on the subordinate's personality traits and growth image shown in Domains 1442 and 1443, examples of effective ways to start the interview or how to speak to them are presented. For example, for a subordinate with trait A (concentration), it is suggested to use considerate words such as, "I feel like you're taking on too much because you have such a strong sense of responsibility." For exploring solutions, based on the perspectives presented in Domain 1444, specific examples of questions or approaches are presented on how to ask questions and explore solutions together, taking into account the subordinate's personality and growth image. For example, for a subordinate with trait C (discrimination), questions that clarify specific problems are suggested, such as, "What is the one thing you want to improve most in the current situation?" In terms of specific actions, hints on what concrete actions can be taken based on the discussion, or points to consider when encouraging action, will be presented. For closing, based on the subordinate's characteristics, growth vision, and the content of the discussion, appropriate remarks to conclude the meeting effectively, or suggestions for the next follow-up, will be made. For example, for a subordinate with characteristic D (diffusiveness) who desires "relentless" growth, it may be suggested to use encouraging words such as, "I think the ideas you shared today are interesting! Let's try them out right away."

[0155] As described above, in this use case, server 20A first acquires "evaluation information" regarding the employee's concerns and "information about the employee's thinking" (personality traits or growth image, etc.) obtained through dialogue with the AI ​​mentor. Next, based on this multifaceted information, the service processing module 2033 takes the lead in instructing the generating AI system 30 to generate several "first perspectives" that should be focused on during the interview. Subsequently, by combining these "first perspectives" and "information about thinking," the system generates "interview support information" that is individually optimized for each employee, showing how to conduct the interview (including how to start the interview, how to explore solutions, how to encourage action, and how to close the interview). The generated "interview support information" is then stored so that it can be presented to the employee's supervisor for use in preparing for or conducting a one-on-one interview. In this way, by providing supervisors with specific guidelines or hints for questions based on detailed information about each employee, the generating AI can avoid the superficial and uninspiring answers that often occur when the generating AI simply presents general solutions. As a result, supervisors can conduct more effective and high-quality one-on-one interviews that are deeply tailored to the characteristics and circumstances of each employee. Furthermore, the interview support in the application examples of this invention can address the challenge of individualized communication that responds to the diverse personalities and values ​​of employees, which has become difficult as organizations have grown in size and values ​​have diversified. In addition, by effectively utilizing the personality traits and values ​​of subordinates as data and supporting communication that is not uniform but tailored to the individual's psychological state, it is possible to reduce the risk of decreased engagement and employee turnover.

[0156] <Basic Computer Hardware Configuration> Figure 16 is a block diagram showing the basic hardware configuration of computer 90. Computer 90 comprises at least a processor 901, main memory 902, auxiliary storage 903, and a communication interface IF991. These are electrically connected to each other by a communication bus.

[0157] The processor 901 is hardware for executing the instruction set written in a program. The processor 901 consists of an arithmetic unit, registers, peripheral circuits, etc.

[0158] Main memory 902 is used to temporarily store programs and data processed by programs, etc. For example, it is a volatile memory such as DRAM (Dynamic Random Access Memory).

[0159] Auxiliary storage device 903 refers to a storage device for saving data and programs. Examples include flash memory, HDD (Hard Disc Drive), magneto-optical disk, CD-ROM, DVD-ROM, and semiconductor memory.

[0160] The IF991 communication interface is an interface for inputting and outputting signals for communication with other computers via a network using wired or wireless communication standards. A network consists of various mobile communication systems, such as the internet, LANs, and wireless base stations. For example, a network includes 3G, 4G, and 5G mobile communication systems, LTE (Long Term Evolution), and wireless networks that can connect to the internet via designated access points (e.g., Wi-Fi®). When connecting wirelessly, communication protocols include, for example, Z-Wave®, ZigBee®, and Bluetooth®. When connecting via a wired connection, the network also includes connections made directly via USB (Universal Serial Bus) cables, etc.

[0161] Furthermore, by distributing all or part of each hardware configuration across multiple computers 90 and connecting them to each other via a network, a computer 90 can be virtually realized. Thus, the concept of computer 90 includes not only a computer 90 housed in a single enclosure or case, but also a virtualized computer system.

[0162] <Basic Functional Configuration of Computer 90> The functional configuration of the computer realized by the basic hardware configuration of computer 90 (Figure 16) will be explained. The computer comprises at least one functional unit: a control unit, a memory unit, and a communication unit.

[0163] Furthermore, the functional units of computer 90 can also be realized by distributing all or part of each functional unit across multiple computers 90 interconnected via a network. The concept of computer 90 includes not only a single computer 90 but also a virtualized computer system.

[0164] The control unit is realized when the processor 901 reads various programs stored in the auxiliary storage device 903, loads them into the main memory device 902, and executes processing according to those programs. The control unit can realize various functional units that perform information processing depending on the type of program. In this way, the computer is realized as an information processing device that performs information processing.

[0165] The memory unit is implemented by the main memory 902 and the auxiliary memory 903. The memory unit stores data, various programs, and various databases. The processor 901 can also reserve memory areas corresponding to the memory unit in the main memory 902 or the auxiliary memory 903 according to the program. The control unit can also cause the processor 901 to perform operations such as adding, updating, and deleting data stored in the memory unit according to the various programs.

[0166] A database, specifically a relational database, is used to manage and link together tabular data sets called masters, which are structurally defined by rows and columns. In a database, tables are called tables, masters are called masters, the columns of tables are called columns, and the rows of tables are called records. In a relational database, relationships can be established and linked between tables and masters. Typically, each table and master has a primary key column to uniquely identify records, but setting a primary key column is not mandatory. The control unit can instruct the processor 901 to add, delete, or update records in specific tables and masters stored in the memory unit, according to various programs. Furthermore, by storing data, various programs, and various databases in the memory unit, the information processing device and information processing system related to this disclosure can be considered to have been manufactured.

[0167] Furthermore, the databases and masters in this disclosure may include any data structures (lists, dictionaries, associative arrays, objects, etc.) in which information is structurally defined. Data structures also include data that can be considered as data structures by combining data with functions, classes, methods, etc. written in any programming language.

[0168] The communication unit is implemented by the communication IF991. The communication unit provides the functionality to communicate with other computers 90 via the network. The communication unit can receive information transmitted from other computers 90 and input it to the control unit. The control unit can cause the processor 901 to perform information processing on the received information according to various programs. The communication unit can also transmit information output from the control unit to other computers 90.

[0169] Furthermore, each of the above-mentioned configurations, functions, processing units, processing means, etc., may be implemented in hardware, in whole or in part, for example, by designing them as integrated circuits. The present invention can also be implemented by software program code that realizes the functions of the embodiment. In this case, a storage medium on which the program code is recorded is provided to a computer, and the processor of that computer reads the program code stored in the storage medium. In this case, the program code read from the storage medium itself realizes the functions of the embodiment described above, and the program code itself and the storage medium on which it is stored constitute the present invention. Examples of storage media used to supply such program code include flexible disks, CD-ROMs, DVD-ROMs, hard disks, SSDs, optical disks, magneto-optical disks, CD-Rs, magnetic tapes, non-volatile memory cards, ROMs, etc.

[0170] Furthermore, the program code that implements the functions described in this embodiment can be implemented in a wide range of programming or scripting languages, such as assembler, C / C++, Perl, Shell, PHP, and Java (registered trademark).

[0171] Furthermore, the program code for the software that implements the functions of the embodiment may be distributed via a network and stored in a storage means such as a computer's hard disk or memory, or in a storage medium such as a CD-RW or CD-R, and the computer's processor may read and execute the program code stored in the storage means or storage medium.

[0172] The functions realized by the components described herein may be implemented in a circuit or processing circuitry, including general-purpose processors, application-specific processors, integrated circuits, ASICs (Application Specific Integrated Circuits), CPUs (a Central Processing Unit), conventional circuits, and / or combinations thereof, programmed to realize the functions described herein. A processor is considered a circuit or processing circuitry, including transistors and other circuits. A processor may be a programmed processor that executes a program stored in memory. In this specification, circuitry, unit, and means are hardware programmed to perform or execute the functions described herein. Such hardware may be any hardware disclosed herein, or any hardware known to be programmed to perform or execute the functions described herein. If the hardware is a processor that is considered to be a type of circuitry, then the circuitry, means, or unit is a combination of hardware and software used to constitute the hardware and / or processor.

[0173] While several embodiments of this disclosure have been described above, these embodiments can be implemented in a variety of other forms, and various omissions, substitutions, and modifications are permitted without departing from the spirit of the invention. These embodiments and their variations are included in the scope and spirit of the invention, as well as in the claims and their equivalents.

[0174] (Note) The details described in each of the above embodiments are noted below.

[0175] (Note 1) A program for operating a computer that includes a processor and memory, The program above is directed to the processor, To initiate the user interaction phase, trigger information that triggers the start of the interaction is input into the generating AI system, and an initial comment is obtained from the generating AI system, which is adjusted to elicit personal information from the user based on a basic prompt definition containing comprehensive instructions for interaction control and the trigger information, and presented to the user. In the information gathering phase of the dialogue, the user inputs a query into the generating AI system, and based on the basic prompt definition and the query, the generating AI system obtains a response to extract information from the user regarding a predetermined viewpoint and presents it to the user. The generation AI system, when it determines that predetermined termination requirements have been met based on the basic prompt definition and the content of the dialogue, obtains a comment from the generation AI system prompting the user to confirm that the dialogue should be terminated, and presents it to the user. In the closing phase of the dialogue, the user is informed that they have agreed to the end of the dialogue, and the user's evaluation information is obtained from the generating AI system based on the basic prompt definition and the content of the dialogue. The steps include presenting the aforementioned evaluation information to the user, A program that executes something. (Note 2) The predetermined termination requirements include the program described in (Appendix 1), wherein the content of the dialogue has been confirmed by the user for all items of the predetermined perspective, and at least one episode has been obtained from the user for each of the perspectives, or the user has requested to terminate the dialogue. (Note 3) The evaluation information includes an evaluation of the predetermined viewpoint from which the generating AI system collected information during the information collection phase, and the evaluation is a program as described in (Appendix 1) or (Appendix 2), which includes text describing the evaluation and indicators. (Note 4) The basic prompt definition includes information for setting the tone of the comments output by the generating AI system in response to the user's specifications, as described in any of the programs in (Appendix 1) to (Appendix 3). (Note 5) The basic prompt definition includes an instruction to cause the generating AI system to output a summary of the content of the dialogue that prompts the user to reflect on the content of the dialogue, based on the content of the dialogue, during the closing phase, as described in any of (Appendix 1) to (Appendix 4). (Note 6) The evaluation information includes advice regarding the user's future actions, and such advice includes information regarding the user's next actions, as described in any of the programs described in (Appendix 1) to (Appendix 5). (Note 7) The program according to any one of (Appendix 1) to (Appendix 6), wherein in the information gathering phase, the processor further performs the step of inputting relevant information extracted from an internal information database storing internal company information into the generating AI system in response to an input query from the user or a request from the generating AI system. (Note 8) The program according to any one of (Appendix 1) to (Appendix 7), wherein in the closing phase, if the evaluation of any of the predetermined viewpoints among the user evaluation information obtained from the generating AI system meets a predetermined threshold value, the processor further performs the step of reporting the existence of the user to a predetermined reporting destination. (Note 9) The program described in (Appendix 8), which causes the processor to further perform the step of presenting the evaluation information to the predetermined reporting destination in response to a request from the predetermined reporting destination. (Note 10) The program according to any one of (Appendix 1) to (Appendix 9), which causes the processor to further perform the step of inputting the user's evaluation information into a pre-trained model that has been trained to estimate indicators related to retirement, and obtaining the user's indicators related to retirement from the pre-trained model. (Note 11) The program described in (Appendix 10), which causes the processor to further perform the step of reporting to a predetermined reporting destination that the acquired retirement indicators meet predetermined requirements, the existence of a user whose retirement indicators meet the predetermined requirements. (Note 12) An information processing apparatus comprising a processor and memory, wherein the processor executes all steps in any of the programs described in (Appendix 1) to (Appendix 11). (Note 13) A method to be executed on a computer comprising a processor and memory, characterized in that the processor executes all steps in any of the programs described in (Appendix 1) to (Appendix 11). (Note 14) A system characterized by comprising means for executing all steps in the program described in any of (Appendix 1) to (Appendix 11). [Explanation of Symbols]

[0176] 1... System 10…Terminal device 12…Communication IF 13…Input device 14…Output device 15…Memory 16…Storage 19… Processor 20... Server 20A…Server 22...Communication IF 23…Input / Output Interface 25…Memory 2 hours… storage 29… Processor 30…Generating AI system 40…Internal Information Database 50…Talent Management System 80…Network

Claims

1. A program for operating a computer that includes a processor and memory, The program above is directed to the processor, To initiate the user interaction phase, trigger information that triggers the start of the interaction is input into the generating AI system, and an initial comment is obtained from the generating AI system, which is adjusted to elicit personal information from the user based on a basic prompt definition containing comprehensive instructions for interaction control and the trigger information, and presented to the user. In the information gathering phase of the dialogue, the user inputs a query into the generating AI system, and based on the basic prompt definition and the query, the AI ​​system controls the dialogue to deeply empathize with the user's words and provide a sense of security. The AI ​​system then obtains responses from the generating AI system to elicit information, along with specific episodes, that will reveal the user's true feelings and clarify the details of their worries or negative factors from a predetermined perspective, and presents these responses to the user. The generation AI system, when it determines that predetermined termination requirements have been met based on the basic prompt definition and the content of the dialogue, obtains a comment from the generation AI system prompting the user to confirm that the dialogue should be terminated, and presents it to the user. In the closing phase of the dialogue, the user is informed that the user has agreed to the end of the dialogue, and the user's evaluation information is obtained from the generating AI system based on the basic prompt definition and the content of the dialogue. The steps include presenting the aforementioned evaluation information to the user, A program that performs the steps of inputting the user's evaluation information into a pre-trained model that has been trained to estimate indicators related to retirement, and obtaining the user's retirement indicators from the trained model.

2. The program according to claim 1, wherein the predetermined termination condition includes the user confirming all items of the predetermined viewpoint and obtaining at least one episode from the user for each viewpoint, or the user requesting to terminate the dialogue.

3. The program according to claim 1, wherein the evaluation information includes an evaluation of the predetermined viewpoint from which the generating AI system has collected information in the information collection phase, and the evaluation includes text describing the evaluation and an index.

4. The program according to claim 1, wherein the basic prompt definition includes information for setting the tone of the comments output by the generating AI system in response to the user's specification.

5. The program according to claim 1, wherein the basic prompt definition includes an instruction in the closing phase to cause the generating AI system to output a summary of the content of the dialogue that prompts the user to reflect on the content of the dialogue.

6. The program according to claim 1, wherein the evaluation information includes advice regarding the user's future actions, and such advice includes information regarding the user's next action.

7. The program according to claim 1, wherein in the information gathering phase, the processor further performs the step of inputting relevant information extracted from an internal information database storing internal company information into the generating AI system in response to an input query from the user or a request from the generating AI system.

8. The program according to claim 1, wherein in the closing phase, if the processor finds that any of the evaluations from the predetermined viewpoints among the user evaluation information obtained from the generating AI system meets a predetermined threshold value, the processor further performs the step of reporting the existence of the user to a predetermined reporting destination.

9. The program according to claim 8, further comprising the step of causing the processor to present the evaluation information to the predetermined reporting destination in response to a request from the predetermined reporting destination.

10. The program according to claim 1, wherein the processor further causes the processor to perform the step of reporting to a predetermined reporting destination that there is a user whose retirement indicators satisfy the predetermined requirements, when the acquired retirement indicators satisfy the predetermined requirements.

11. An information processing apparatus comprising a processor and memory, wherein the processor executes all steps in the program described in any one of claims 1 to 10.

12. A method to be performed on a computer comprising a processor and memory, wherein the processor performs all steps in a program according to any one of claims 1 to 10.

13. A system characterized by comprising means for executing all steps in the program described in any one of claims 1 to 10.

Images