Programs, information processing devices, methods, and systems

Abstract

We provide a user-friendly system that utilizes generative AI. [Solution] A program to be executed on a computer having a processor and memory, which stores information about multiple users and multiple pieces of information associated with each of the multiple pieces of information about the multiple users, for explaining predetermined content to the user. The program causes the processor to perform the following steps: present the user with multiple options required by the user and accept the selection of one of the multiple options; refer to the instruction sentence included in the accepted option and information about the user who selected the option and read predetermined information from the multiple pieces of information; input the input instruction sentence and the read predetermined information into a large language model and have the large language model output a response to the input instruction sentence; and present the output response to the user who selected the option.

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, systems using large language models (LLMs) have been provided. A large language model is a language model specialized for natural language processing (NLP) and is a type of so-called generative artificial intelligence (AI). Generally, when an instruction sentence called a prompt is input, an LLM outputs a sentence based on the instruction sentence. 【0003】 For example, Patent Document 1 discloses a dialogue server that responds with search results based on an input by a user. The dialogue server acquires an input keyword from a user terminal operated by the user and acquires user information associated with the user. The dialogue server searches a search database that stores search keywords and response candidates in association with each other for a response candidate corresponding to the input keyword and the user information. The dialogue server transmits the searched response candidate to the user terminal. When the dialogue server receives a negative intention for the response candidate from the user terminal, the dialogue server searches for a response candidate using another search keyword without using the user information. 【Prior Art Documents】 【Patent Documents】 【0004】 【Patent Document 1】 Japanese Patent No. 6975323 【Summary of the Invention】 【Problems to be Solved by the Invention】 【0005】 In Patent Document 1, the dialogue server performs a search using user information to narrow down the response candidates. The dialogue server obtains answers from the FAQ database that correspond to the candidate questions for which it has received affirmative instructions from the user terminal. However, Patent Document 1 does not describe a system that utilizes generative AI. 【0006】 The purpose of this disclosure is to provide a user-friendly system that utilizes generative AI. [Means for solving the problem] 【0007】 To solve the aforementioned problems, a program according to one aspect of the present disclosure is a program to be executed on a computer having a processor and memory, and stores information about a plurality of users and a plurality of pieces of information associated with each of the plurality of users for explaining predetermined content to the user. The program causes the processor to perform the following steps: present a plurality of options required by the user to the user and accept the user's selection of one of the plurality of options; refer to the instruction sentence included in the accepted option and information about the user who selected the option and read predetermined information from the plurality of pieces of information; input the input instruction sentence and the read predetermined information into a large language model and have the large language model output a response to the input instruction sentence; and present the output response to the user who selected the option. [Effects of the Invention] 【0008】 According to this disclosure, it is possible to provide a user-friendly system that utilizes generative AI. [Brief explanation of the drawing] 【0009】 [Figure 1] This is a block diagram showing an example of the overall configuration of System 1. [Figure 2] This is a block diagram showing an example configuration of terminal device 10. [Figure 3]This block shows an example configuration for server 20. [Figure 4] This diagram shows the data structure of employee database 40. [Figure 5] This diagram shows the data structure of the rule database 501. [Figure 6] This diagram shows the data structure of manual database 502. [Figure 7] This flowchart shows an example of server 20's operation. [Figure 8] This is a schematic diagram showing an example of the display 141 screen when the site screen 1411 is displayed. [Figure 9] This is a schematic diagram showing an example of the screen of display 141 when answer 1416 is displayed. [Figure 10] 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】 [1. Overview] In this embodiment, information about multiple users and multiple pieces of information associated with each of the multiple user pieces are stored to explain predetermined content to the user. A user is, for example, an employee of a company. In this embodiment, the information about multiple users and the multiple pieces of information are stored in separate databases. The server according to this embodiment presents the user with multiple options that the user needs and accepts the user's selection of one of the options. The server according to this embodiment reads predetermined information from the multiple pieces of information by referring to the instruction sentence included in the accepted option and the information about the user who selected the option. The server according to this embodiment inputs the instruction sentence and the read predetermined information into the LLM and causes the LLM to output a response to the input instruction sentence. The server according to this embodiment presents the output response to the user who selected the option. 【0012】 [2. Overall System Configuration] Figure 1 is a block diagram showing an example of the overall configuration of System 1. System 1 is a system for providing a service that answers user questions using, for example, LLM (hereinafter referred to as the Generative AI Answer Service). In this specification, the Generative AI Answer Service is explained using the example of a case where the Generative AI Answer Service is introduced to a company and an employee belonging to that company uses the Generative AI Answer Service as a user. However, the example of providing the Generative AI Answer Service is not limited to cases where the user is an employee. 【0013】 System 1 shown in Figure 1 includes, for example, a terminal device 10, a server 20, a generation AI system 30, an employee database 40, and a RAG database 50. The terminal device 10, server 20, generation AI system 30, employee database 40, and RAG database 50 are connected to each other via, for example, a network 80. 【0014】 Figure 1 shows an example where System 1 includes one terminal device 10, for the sake of illustration simplification. However, in reality, System 1 includes a number of terminal devices 10 equal to the number of employees. 【0015】 In FIG. 1, an example in which the system 1 includes one server 20 is shown. However, for example, an aggregate of a plurality of devices may be regarded as one server 20. The way of distributing the plurality of functions required to implement the server 20 to one or more hardware can be appropriately determined according to the processing capacity of each hardware and / or the specifications required for the server 20, etc. 【0016】 In FIG. 1, an example in which the system 1 includes one generative AI system 30 is shown. However, the number of generative AI systems 30 included in the system 1 may be two or more. Also, in FIG. 1, an example in which the generative AI system 30 is independent of the server 20 is shown. However, the server 20 may include the functions of the generative AI system 30. That is, the server 20 may store the LLM included in the generative AI system 30. 【0017】 In FIG. 1, an example in which the employee database 40 and the RAG database 50 are independent of each other is shown. However, for example, both of them may be integrated into one database. Also, for example, a management table (not shown) may manage the employee database 40 and the RAG database 50 in association with each other. 【0018】 In FIG. 1, an example in which each of the employee database 40 and the RAG database 50 is independent of the server 20 is shown. However, the server 20 may include at least one of the functions of the employee database 40 and the RAG database 50. That is, at least one of the employee database 40 and the RAG database 50 may be constructed in the server 20. 【0019】 The terminal device 10 is, for example, an information processing device operated by an employee who is a user, and is provided to each individual employee. The terminal device 10 is realized by, for example, a mobile terminal such as a smartphone or a tablet. In the present embodiment, it is assumed that the terminal device 10 is a tablet. The terminal device 10 may also be realized by, for example, a stationary PC (Personal Computer), a laptop PC, etc. 【0020】 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 for receiving input operations from the user (e.g., a touch panel, touchpad, pointing device such as a mouse, keyboard, etc.). The output device 14 is a device for presenting information to employees (display, speaker, etc.). In this embodiment, the terminal device 10 is assumed to have a touch panel that integrates the input device 13 and the output device 14. 【0021】 Server 20 is, for example, an information processing device for managing and operating a generating AI response service, and is implemented by a computer connected to the network 80. Server 20 is, for example, an API (Application Programming Interface) server. As shown in Figure 1, Server 20 includes a communication IF 22, an input / output IF 23, memory 25, storage 26, and a processor 29. The input / output IF 23 functions as an interface for an input device to receive input operations from the administrator of the generating AI response service, and an output device to output information to the administrator. 【0022】 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. 【0023】 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 【0024】 The generation AI system 30 inputs text data sent from the server 20 into the LLM and causes the LLM to output a response based on the input text data. The generation AI system 30 then sends the response output from the LLM back to the server 20. 【0025】 The employee database 40 is a database that stores employee information of multiple employees. Employee information is information about an employee and is an example of "user information" in one aspect of this disclosure. Employee information includes, for example, an employee's personal information and information about the terminal device 10 provided to the employee (hereinafter referred to as terminal information). Personal information may be obtained, for example, from personnel information stored on the management server or management terminal (both not shown) of the employee's company, or from information from a pre-survey conducted with the employee. Terminal information may be obtained, for example, from the purchase history of the terminal device 10 stored on the management server or management terminal, or by the server 20 receiving information input from the employee. Employee information may refer to information used in other services. For this reason, the employee database 40 may be built on a server different from the server 20, for example. Alternatively, the employee database 40 may be provided on the server 20, for example. 【0026】 The RAG database 50 is a database that stores descriptive information that is searchable by RAG (Retrieval-Augmented Generation) executed by the server 20. The descriptive information is information associated with employee information, intended to explain predetermined content to the employee to whom the employee information pertains. The descriptive information includes, for example, various company rules (e.g., employment regulations), various manuals (e.g., operation manual for terminal device 10), etc. Multiple pieces of descriptive information stored in the RAG database 50 are an example of "multiple pieces of information" relating to one aspect of this disclosure. 【0027】 Each information processing device, such as the terminal device 10, the server 20, and the generation AI system 30, is composed of a computer 90 (see Figure 10) equipped with an arithmetic unit and a memory device. The basic hardware configuration of the computer 90 and the basic functional configuration of the computer 90 realized by this basic hardware configuration will be described later. Note that explanations of the terminal device 10, the server 20, and the generation AI system 30 that overlap with the basic hardware configuration of the computer 90 and the basic functional configuration of the computer will be omitted. 【0028】 <2-1 Configuration of Terminal Devices> Figure 2 is a block diagram showing an example configuration of the terminal device 10 shown in Figure 1. As shown in Figure 2, the terminal device 10 includes a communication unit 120, an input device 13, an output device 14, an audio processing unit 170, a microphone 171, a speaker 172, a camera 160, a position information sensor 150, an acceleration sensor 155, 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. 【0029】 The communication unit 120 performs processing such as modulation and demodulation processing for the terminal device 10 to communicate with an external device (for example, a server 20). The communication unit 120 performs transmission processing on the signal generated by the control unit 190 and transmits it to the external device. The communication unit 120 performs reception processing on the signal received from the external device and outputs it to the control unit 190. 【0030】 The input device 13 is a device for employees 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 or the like, the input device 13 may be implemented by a reader, keyboard, mouse, etc. The input device 13 converts the instructions input by the employee into electrical signals and outputs them 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. 【0031】 The output device 14 is a device for presenting information to employees. The output device 14 is implemented, for example, by a display 141. The display 141 displays various information according to the control of the control unit 190. The display 141 is implemented, for example, by an LCD (Liquid Crystal Display) or an organic EL (Electro-Luminescence) display. 【0032】 The audio processing unit 170 performs, for example, digital-to-analog conversion processing of the audio signal. The audio processing unit 170 converts the signal provided from the microphone 171 into a digital signal and provides the converted signal to the control unit 190. The audio processing unit 170 also provides the audio signal to the speaker 172. The audio processing unit 170 is implemented, for example, by an audio processing processor. The microphone 171 receives an audio input and provides the audio signal corresponding to that audio input to the audio processing unit 170. The speaker 172 converts the audio signal provided from the audio processing unit 170 into audio and outputs the audio to the outside of the terminal device 10. 【0033】 Camera 160 is an imaging device that captures images using visible light. In other words, camera 160 is a device that receives visible light using a photodetector and outputs image data as a shooting signal. Camera 160 captures subjects in a certain direction and within a certain shooting range relative to the terminal device 10 and outputs image data as a result of the capture. If camera 160 has a function that allows adjustment of the shooting range, or more precisely, the angle of view, camera 160 also outputs information regarding this angle of view. Such a function is called a zoom function. 【0034】 The location information sensor 150 is a sensor that detects the position of the terminal device 10, and is generally a GNSS device, such as a GPS 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 based on the received signals, the current position of the terminal device 10, which is equipped with a GPS module, is detected in coordinate values. The location information sensor 150 may also detect the current position of the terminal device 10 from the position of a wireless base station to which the terminal device 10 is connected via the communication unit 120. 【0035】 The acceleration sensor 155 is a sensor that detects the acceleration applied to the terminal device 10. Preferably, the acceleration sensor 155 has the function of detecting the tilt around each axis (X axis, Y axis, Z axis) of a three-dimensional coordinate system with the position of the terminal device 10 as the origin. An acceleration sensor 155 having such a function can detect the orientation of the terminal device 10, that is, its direction with respect to the X, Y, and Z axes, by detecting the gravitational acceleration of the Earth's gravity. 【0036】 The storage unit 180 is implemented by the memory 15 and storage 16 shown in Figure 1, and stores data and programs used by the terminal device 10. The programs include, for example, a web browser application. 【0037】 The control unit 190 is implemented, for example, by the processor 19 reading a program (including application 2021) stored in the memory unit 180 and executing the instructions contained in the program. The control unit 190 controls the operation of the terminal device 10. By operating according to the loaded 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. 【0038】 The operation reception unit 191 performs processing to receive instructions or information input from the input device 13. Specifically, the operation reception unit 191 receives instructions or information input from the touch-sensitive device 131. 【0039】 The transmitting / receiving unit 192 performs processing to enable the terminal device 10 to send and receive data with an external device in accordance with a communication protocol. Specifically, the transmitting / receiving unit 192 transmits instructions or information input by the user to the server 20. The transmitting / receiving unit 192 receives information transmitted from the server 20. 【0040】 The presentation control unit 193 controls the output device 14 to present various information to the user. For example, the presentation control unit 193 presents the user with the response output from the LLM included in the generation AI system 30. 【0041】 <2-2 Server Configuration> Figure 3 is a block diagram showing an example 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. 【0042】 The communication unit 201 performs processing to enable the server 20 to communicate with external devices (for example, terminal device 10, generation AI system 30). 【0043】 The storage unit 202 is implemented by memory 25 and storage 26, and stores data and programs used by the server 20. The programs include application programs such as web browser applications. 【0044】 App 2021 is application software for running and managing a generated AI response service. Users can access App 2021 on the server 20, for example, by using a web browser application installed on terminal device 10. App 2021 may also be stored on terminal device 10, for example. In this case, App 2021 runs in the background of other applications installed on terminal device 10, for example. 【0045】 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 that program. The control unit 203 controls the operation of the server 20. By operating according to the loaded program, the control unit 203 performs the functions of the receive control module 2031, the transmit control module 2032, the presentation control module 2033, and the question processing module 2034. 【0046】 The receive control module 2031 controls the process by which the server 20 receives signals from an external device according to a communication protocol. The transmit control module 2032 controls the process by which the server 20 transmits signals to an external device according to a communication protocol. 【0047】 The presentation control module 2033 controls the process of presenting various types of information to the user. For example, the presentation control module 2033 controls the presentation control unit 193 to present the user with the answers output from the LLM included in the generation AI system 30. 【0048】 The question processing module 2034 presents employees with multiple options and accepts their selection from among them. The options may include information necessary for employees to carry out general activities related to their company, and may also include information that employees are presumed to want to know at present. Specific examples of options include options containing information necessary for employees to perform their assigned tasks (e.g., how to use terminal device 10, how to use project management tools), options containing information necessary for general internal activities (e.g., how to use internal information sharing tools, how to connect to the internal network, information about recreation), and options containing information necessary for general external activities (e.g., how to use the cloud-based telephone system). 【0049】 In this embodiment, the question processing module 2034 presents employees with multiple options, for example, through the site screen of the company's internal portal site. In other words, in this embodiment, the question processing module 2034 implements a function that provides employees with the information they need on the company's internal portal site. The company's internal portal site is, for example, a portal site for employees to obtain various information within the company, and multiple employees (specifically, all employees) can access it simultaneously. The company's internal portal site presents employees with multiple options on the site screen, for example, based on the employee information of the employee who accessed it. 【0050】 More specifically, the question processing module 2034 controls, for example, the presentation control unit 193 to display a site screen containing multiple options on the display 141. Note that the manner in which multiple options are presented to employees is not limited to the examples in this embodiment. 【0051】 In this embodiment, the question processing module 2034 accepts a selection operation in which an employee selects a specific option from among multiple options displayed on a site screen, for example. Specifically, the question processing module 2034 accepts the aforementioned selection operation by receiving selection information regarding a specific option transmitted from the terminal device 10. Note that the manner in which the selection of an option is accepted is not limited to the example of this embodiment. 【0052】 The question processing module 2034 refers to the instruction text included in the accepted options and the employee information of the employee who selected the option, and retrieves the required information from among multiple explanatory pieces of information. 【0053】 "Instructional statements included in the options" are, for example, sentences that can be interpreted as instructing the reading of specific explanatory information contained in the text information that constitutes at least part of the options. For example, if an option selected by an employee includes the text information "How to connect to the company network," this text information can be interpreted as instructing the reading of the operation manual for the terminal device 10 that will be connected to the company network. 【0054】 In this embodiment, the question processing module 2034, for example, refers to the selection information (which includes various information about the employee and various information about the terminal device 10) received from the terminal device 10 to identify employee information corresponding to the selection information from among a plurality of employee information stored in the employee database 40. The question processing module 2034, for example, refers to the selection information received from the terminal device 10 and the identified employee information to read predetermined information from among a plurality of descriptive information stored in the RAG database 50. In other words, the question processing module 2034 performs information retrieval in RAG using the selection information received from the terminal device 10 and the identified employee information as queries. 【0055】 The specified information is, for example, explanatory information that is highly relevant to the instructions in the instruction statement included in the selected information, among multiple pieces of explanatory information. For example, in the case of selected information regarding the option "How to connect to the company network," the operation manual for terminal device 10 is read out as the specified information from among multiple pieces of explanatory information. 【0056】 There are no particular limitations on the information retrieval methods performed by the question processing module 2034 in RAG. For example, the question processing module 2034 retrieves predetermined information from the RAG database 50 by performing a keyword search or semantic search based on selected information and employee information. 【0057】 Furthermore, if no suitable option is displayed on the site screen, employees may enter a free-text instruction, for example, to perform an information search in RAG. In other words, the question processing module 2034 may accept a free-text instruction from an employee. In this case, the question processing module 2034 may, for example, refer to the accepted instruction and identify the employee information corresponding to that instruction from among the multiple employee information stored in the employee database 40. 【0058】 The question processing module 2034 inputs the input instruction and the predetermined information it reads into the LLM, and causes the LLM to output a response to the input instruction. 【0059】 In this embodiment, the question processing module 2034 transmits, for example, selection information (including instruction sentences included in the choices) received from the terminal device 10 and predetermined information read from the RAG database 50 to the generation AI system 30. The generation AI system 30 inputs, for example, the instruction sentence included in the selection information received from the server 20 as a prompt to the LLM, and also inputs the predetermined information received from the server 20 to the LLM. Here, for example, the predetermined information and the prompt may be input to the LLM individually, or a prompt with the predetermined information incorporated may be input to the LLM. The generation AI system 30 obtains, for example, the answer to the prompt (instruction sentence) from the LLM and transmits it to the server 20. The reception control module 2031 receives, for example, the answer transmitted from the generation AI system 30. 【0060】 As a result, the question processing module 2034 performs answer generation in RAG based on the selection information received from the terminal device 10 and predetermined information read from the RAG database 50. 【0061】 Furthermore, if the question processing module 2034 receives, for example, a free-word instruction from an employee, it may send the received free-word instruction to the generation AI system 30 instead of the selected information. In this case, the generation AI system 30 may, for example, input the free-word instruction received from the server 20 as a prompt to the LLM. 【0062】 [3 Data Structure] Figures 4-6 show the data structure of the database in System 1. Note that Figures 4-6 are merely examples and do not exclude any data not shown. Furthermore, even data within the same database may be stored in separate memory areas. 【0063】 Figure 4 shows the data structure of the employee database 40. The employee database 40 shown in Figure 4 is a table that has columns for personal information and terminal information, with employee ID as the key. The various types of information stored in the employee database 40 are pre-configured by the administrator, for example, and edited and updated as needed. 【0064】 The "Employee ID" field is, for example, a field that stores identification information (identifier) ​​to uniquely identify each of the multiple employees being managed. The "Personal Information" field is, for example, a field that stores personal information for each of the multiple employees. Personal information, for example, constitutes at least a part of the employee information. Personal information includes, for example, the employee's name, hiring information, job title, department, and information about the project they are assigned to. The "Terminal Information" field is, for example, a field that stores terminal information for terminal devices 10 provided to each of the multiple employees. Terminal information, for example, constitutes at least a part of the employee information. Terminal information includes, for example, the type of terminal device 10, specifications, and the employee's usage history. 【0065】 Furthermore, the employee database 40 may store information other than the information stored in each of the aforementioned items. In other words, employee information may include information other than the information stored in the employee database 40. Also, from the perspective of improving search accuracy in RAG, the employee database 40 may store personal information and terminal information that are weighted according to each employee's past question trends, past choice trends, etc. 【0066】 In this embodiment, the RAG database 50 consists of, for example, a rule database 501 shown in Figure 5 and a manual database 502 shown in Figure 6. The rule database 501 and the manual database 502 are constructed, for example, independently of each other. 【0067】 The configuration of the RAG database 50 is not limited to the example of this embodiment. For example, the RAG database 50 may be configured such that the rule database 501 and the manual database 502 are associated with an unillustrated management table. Alternatively, the RAG database 50 may be configured as a single integrated database combining the rule database 501 and the manual database 502. Furthermore, the RAG database 50 may consist of only either the rule database 501 or the manual database 502, or it may include databases other than these two. 【0068】 Figure 5 shows the data structure of the rules database 501. The rules database 501 shown in Figure 5 is a table that has columns for employee ID, employee information, rules content, effective date, update date, and target, with rule ID as the key. The various types of information stored in the rules database 501 are pre-configured by the administrator, for example, and edited and updated as needed. 【0069】 The "Rule ID" field stores identification information (identifiers) to uniquely identify various rules of an employee's company, for example. The "Employee ID" field stores identification information (identifiers) to uniquely identify each of multiple employees managed by the employee database 40, for example. The identification information stored in the "Employee ID" field of the rule database 501 is the same as the identification information stored in the "Employee ID" field of the employee database 40, for example. The "Employee Information" field stores employee information for each employee to whom various rules apply. The employee information stored in the "Employee Information" field consists of personal information and terminal information stored in the employee database 40, for example. The "Rule Content" field stores the specific content or outline of various rules, for example. The "Effective Date" field stores the effective date of various rules, for example. The "Update Date" field stores the update date of various rules, for example. The "Target" field is used to record, for example, the job title, department, or duties of employees to whom various regulations apply. 【0070】 As described above, the rules database 501 shown in Figure 5 stores explanatory information such as the specific content or outline of various rules, the effective date, the update date, and the job titles of employees to whom the various rules apply. Furthermore, the rules database 501 shown in Figure 5 stores multiple pieces of explanatory information associated with each employee's information by associating the explanatory information for each type of rule with the employee information to which the respective rule applies. 【0071】 Furthermore, the rule database 501 may store information other than the information stored in each of the aforementioned items. In other words, the explanatory information may include information other than the information stored in the rule database 501. 【0072】 Figure 6 shows the data structure of the manual database 502. The manual database 502 shown in Figure 6 is a table that has columns for employee ID, employee information, manual content, publication date, update date, and target terminal, with the manual ID as the key. The various types of information stored in the manual database 502 are pre-configured by the administrator, for example, and edited and updated as needed. 【0073】 The item "Manual ID" is an item that stores identification information (identifier) ​​to uniquely identify various manuals related to terminal devices 10 owned by the company to which the employee belongs. The item "Employee ID" is an item that stores identification information (identifier) ​​to uniquely identify each of multiple employees who are managed by the employee database 40. The identification information stored in the item "Employee ID" in the manual database 502 is the same as the identification information stored in the item "Employee ID" in the employee database 40. The item "Employee Information" is an item that stores employee information for each employee who can use each of the various manuals. The employee information stored in the item "Employee Information" consists of, for example, personal information and terminal information stored in the employee database 40. The item "Manual Content" is an item that stores, for example, the specific content or summary of various manuals. The item "Issue Date" is an item that stores, for example, the issue date of various manuals. The item "Update Date" is an item that stores, for example, the update date of various manuals. The item "Target Terminal" is, for example, an item that stores the types of terminal devices 10 to which various manuals apply. 【0074】 As described above, the manual database 502 shown in Figure 6 stores, for example, the specific content or summary of various manuals, the publication date, the update date, and the types of terminal devices 10 to which the various manuals apply as descriptive information. Furthermore, the manual database 502 shown in Figure 6 stores, for example, employee information for each employee who can use the various manuals in association with the descriptive information for each manual, thereby storing multiple descriptive pieces of information associated with each employee's information. 【0075】 Furthermore, the manual database 502 may store information other than the information stored in each of the aforementioned items. In other words, the explanatory information may include information other than the information stored in the manual database 502. 【0076】 In the examples in Figures 5 and 6, the rules database 501 and the manual database 502 each store employee information, but the manner in which employee information is associated with multiple explanatory information is not limited to the examples in Figures 5 and 6. For example, a management table may be constructed with employee ID as the key and having columns for employee information, rules ID, and manual ID. The management table may associate employee information with multiple explanatory information by, for example, associating the rules database 501 and the manual database 502 with the employee database 40. In this case, for example, the fields "Employee ID" and "Employee Information" in the rules database 501 and the manual database 502 may be omitted. 【0077】 [4 actions] Referring to Figure 7, an example of the operation of Server 20 when presenting the response output from LLM will be explained. Figure 7 is a flowchart of an example of the operation of Server 20 when presenting the response output from LLM. In this example, it is assumed that a specific employee has logged in from the login screen displayed on Display 141 and that the various functions of the AI ​​response generation service are available. 【0078】 In step S11 shown in Figure 7, the server 20 presents the employee with multiple options and accepts the employee's selection of one of those options. 【0079】 Specifically, for example, the operation reception unit 191 receives a request from an employee to display a site screen. The transmission / reception unit 192 transmits the received display request to the server 20. The question processing module 2034 receives a display request from the terminal device 10 and transmits information to the terminal device 10 to display the site screen. The transmission / reception unit 192 receives information to display the site screen transmitted from the server 20. The presentation control unit 193 transmits the information to display the site screen received from the server 20 to the output device 14, causing the site screen containing multiple options to be displayed on the display 141. As a result, the server 20 presents the employee with multiple options. 【0080】 The operation reception unit 191 receives a selection operation, for example, when a user selects a specific option from among multiple options included in the site screen displayed on the display 141. The transmission / reception unit 192 transmits selection information regarding the specific option to the server 20, for example. The reception control module 2031 receives the selection information transmitted from the terminal device 10, for example. As a result, the server 20 receives the employee's selection of one of the multiple options. 【0081】 The following describes an example of the display 141 when the site screen 1411 is displayed, referring to Figure 8. Figure 8 is a schematic diagram showing an example of the display 141 when the site screen 1411 is displayed. 【0082】 In the example screen shown in Figure 8, for example, the site screen 1411 of the company's internal portal site is displayed on display 141. Area 14111 of the site screen 1411 displays several options 1412 that are necessary (or presumed to be necessary) for employees to carry out all activities related to their company. 【0083】 In the example screen shown in Figure 8, multiple items 1413, categorized according to their respective contents such as activities, tasks, and equipment, are set on the site screen 1411. Multiple options 1412 corresponding to each of the multiple items 1413 are displayed in a list directly below the item 1413. For example, directly below the item 1413 for "Company Equipment and Facilities," options 1412 such as "Reception System" and "Use of Communication Portal" are displayed in a list. 【0084】 Please note that the screen example shown in Figure 8 is merely one example, and various variations are possible regarding the structure and display of site screen 1411, as well as the content and display of options 1412. 【0085】 In step S12, the server 20 refers to the instruction statement included in the accepted options and the employee information of the employee who selected the option, and reads the predetermined information from among the multiple explanatory pieces of information. 【0086】 Specifically, for example, the question processing module 2034 refers to the selection information received from the terminal device 10 and identifies the employee information corresponding to the selection information from among multiple employee information stored in the employee database 40. The question processing module 2034 refers to the selection information received from the terminal device 10 and the identified employee information and reads predetermined information from among multiple explanatory information stored in the rule database 501. Also, for example, the question processing module 2034 refers to the selection information received from the terminal device 10 and the identified employee information and reads predetermined information from among multiple explanatory information stored in the manual database 502. 【0087】 In step S13, the input instruction and the read predetermined information are input to the LLM, and the LLM outputs a response to the input instruction. 【0088】 Specifically, for example, the question processing module 2034 transmits, for example, selection information received from the terminal device 10 and predetermined information read from the rule database 501 or manual database 502 to the generation AI system 30. The generation AI system 30 inputs, for example, an instruction sentence included in the selection information received from the server 20 as a prompt to the LLM, and also inputs the predetermined information received from the server 20 to the LLM. As a result, the server 20 inputs the input instruction sentence and the read predetermined information to the LLM. 【0089】 The generation AI system 30, for example, obtains a response to a prompt (instruction) by having it output from the LLM and sends it to the server 20. The receiving control module 2031 receives the response sent from the generation AI system 30. As a result, the server 20 causes the LLM to output a response to the input instruction. 【0090】 In step S14, the server 20 presents the outputted response to the employee who selected the option. 【0091】 Specifically, for example, the presentation control module 2033 transmits information for displaying the response received from the generation AI system 30 to the terminal device 10. The transmitting / receiving unit 192 receives information for displaying the response transmitted from, for example, the server 20. The presentation control unit 193 transmits the information for displaying the response received from the server 20 to the output device 14, causing the response to be displayed on the display 141. If the output device 14 is a printer, the presentation control unit 193 may, for example, print out a paper medium with the response printed on it from the output device 14. 【0092】 The following describes an example of the screen of display 141 when answer 1416 is displayed, referring to Figure 9. Figure 9 is a schematic diagram showing an example of the screen of display 141 when answer 1416 is displayed. 【0093】 In the example screen shown in Figure 9, for example, the area 14112 of the display 141 shows the option 1414 selected by the employee and employee information 1415 related to option 1414. In this case, the presentation control module 2033, for example, refers to the selection information related to option 1414 received from the terminal device 10 and reads the employee information 1415 corresponding to option 1414 from the employee database 40. The presentation control module 2033, for example, controls the presentation control unit 193 to display the option 14141 indicated by the selection information received from the terminal device 10 and the employee information 1415 read from the employee database 40 on the display 141. 【0094】 In the example screen shown in Figure 9, for example, answer 1416 is displayed in area 14113 of display 141. In this case, the question processing module 2034 executes RAG based on the options 1414 and employee information 1415 displayed in area 14112, for example, and causes the LLM to output answer 1416. The presentation control module 2033 controls the presentation control unit 193, for example, to display the answer 1416 output from the LLM in area 14113. 【0095】 It should be noted that the screen example shown in Figure 9 is merely an example, and various variations are possible regarding the display content, display position, and display manner of answer 1416. For example, it is not necessary to display at least one of the options 1414 and employee information 1415. Also, for example, a field for a follow-up question (which may be in a free-word input format or a selection format from options) for asking a follow-up question about answer 1416, and a field for providing a follow-up answer to the follow-up question, may be displayed on display 141 together with answer 1416. In this case, the question processing module 2034 may, for example, re-identify the employee information based on the content of the follow-up question and re-execute RAG using LLM. 【0096】 [5 Summary] As described above, in this embodiment, the transmitting / receiving unit 192 transmits a request from an employee to display a site screen to the server 20. The question processing module 2034 receives the display request sent from the terminal device 10 and transmits information for displaying the site screen to the terminal device 10. The presentation control unit 193 transmits the information for displaying the site screen received from the server 20 to the output device 14, causing the site screen containing multiple options to be displayed on the display 141. The operation reception unit 191 accepts a selection operation in which the user selects a specific option from among the multiple options included in the site screen displayed on the display 141. The transmitting / receiving unit 192 transmits selection information regarding the specific option to the server 20. The reception control module 2031 receives the selection information transmitted from the terminal device 10. 【0097】 The question processing module 2034 refers to the selection information received from the terminal device 10 and identifies the employee information corresponding to the selection information from among multiple employee information stored in the employee database 40. The question processing module 2034 refers to the selection information received from the terminal device 10 and the identified employee information and reads predetermined information from the rule database 501 or the manual database 502. 【0098】 The question processing module 2034 transmits the selection information received from the terminal device 10 and predetermined information read from the rule database 501 or manual database 502 to the generation AI system 30. The generation AI system 30 inputs the instruction sentence (prompt) contained in the selection information received from the server 20 and the predetermined information into the LLM. The generation AI system 30 obtains the answer to the instruction sentence (prompt) received from the server 20 by having the LLM output and transmits it to the server 20. The reception control module 2031 receives the answer transmitted from the generation AI system 30. The presentation control module 2033 transmits information for displaying the answer received from the generation AI system 30 to the terminal device 10. The presentation control unit 193 transmits information for displaying the answer received from the server 20 to the output device 14 and displays the answer on the display 141. 【0099】 This allows employees to automatically obtain the necessary information as an answer simply by selecting the appropriate option from multiple choices displayed on the site screen. Therefore, it is possible to provide a user-friendly LLM-based system 1 for employees. 【0100】 [6 Variations] <6-1 First Variation> In this embodiment, an example was described in which the server 20 displays multiple options on the site screen of the company's internal portal site. However, the server 20 may, for example, output multiple options from the option creation LLM based on employee information and present them to the employee. 【0101】 Unlike an LLM that outputs a corresponding answer to an option using instructional text and explanatory information included in the options as input data, the LLM for creating options is an LLM for creating multiple options. The LLM for creating options may be included in the generation AI system 30, for example, or in the server 20. In the first modified example, the LLM for creating options is assumed to be included in the generation AI system 30. 【0102】 Specifically, for example, the question processing module 2034 may receive an employee creation request sent from the terminal device 10 and read employee information corresponding to the creation request from the employee database 40. The question processing module 2034 may also send the employee information read from the employee database 40 to the generation AI system 30. 【0103】 The generation AI system 30 may, for example, input employee information received from the server 20 along with a predetermined prompt into the option creation LLM, and have the option creation LLM output multiple options corresponding to the selection request. The generation AI system 30 may, for example, acquire the multiple options output from the option creation LLM and send them to the server 20. 【0104】 The predetermined prompt may be pre-stored in, for example, the memory unit (not shown) of the generation AI system 30, or it may be hardcoded into the program of the generation AI system 30. Alternatively, for example, the predetermined prompt may be entered as free text by the administrator via the input device (not shown) of the server 20. In this case, the server 20 may, for example, send the entered predetermined prompt to the generation AI system 30. There are no particular limitations on the content of the predetermined prompt; for example, it may be an instruction such as, "Please output multiple appropriate options for the employee to input." 【0105】 The question processing module 2034 may, for example, send information for displaying multiple choices received from the generation AI system 30 to the terminal device 10. The presentation control unit 193 may, for example, send information for displaying multiple choices received from the server 20 to the output device 14, causing the multiple choices to be displayed on the display 141. 【0106】 Thus, according to the first modification, the server 20 can, simply by receiving a request from an employee to display multiple options, more flexibly and appropriately present the necessary options to the employee according to the employee information. This makes it possible to provide an LLM-based system that is easier for employees to use. 【0107】 <6-2 Second Variation> In this embodiment, an example was described in which the explanatory information stored in the rule database 501 and manual database 502, which constitute the RAG database 50, is pre-set by the administrator and updated as needed. However, the server 20 may, for example, acquire at least some of the explanatory information (hereinafter referred to as "target explanatory information") from among a plurality of explanatory information and store the acquired target explanatory information in the RAG database 50 in accordance with predetermined rules. 【0108】 Specifically, for example, the receiving control module 2031 may receive target description information transmitted from the management server or management terminal of the employee's company. The management server or management terminal may, for example, send the target description information to the server 20 whenever new target description information is added or whenever the managed target description information is updated. As a result, the server 20 may acquire the target description information. 【0109】 The receiving control module 2031 may, for example, store the target description information received from the management server or management terminal in the RAG database 50 in accordance with predetermined rules. There are no particular limitations on the contents of the predetermined storage. For example, the predetermined rules may include storing the target description information in the rules database 501 if it pertains to company regulations, and storing it in the manual database 502 if it pertains to various manuals. Alternatively, for example, the predetermined rules may include storing the target description information in a field of the RAG database 50 corresponding to the type of company regulations or manuals to which the target description information pertains. 【0110】 More specifically, the receiving control module 2031 may, for example, transmit the subject description information received from the management server or management terminal to the rules database 501 if the subject description information pertains to internal company rules. The rules database 501 may, for example, store the subject description information received from the server 20 in a field corresponding to the type of internal company rule to which the subject description information pertains. Alternatively, for example, the receiving control module 2031 may transmit the subject description information received from the management server or management terminal to the manual database 502 if the subject description information pertains to various manuals. The manual database 502 may, for example, store the subject description information received from the server 20 in a field corresponding to the type of manual to which the subject description information pertains. 【0111】 Thus, according to this second modification, the administrator can pre-set the target descriptive information in the RAG database 50 and avoid the trouble of updating it as needed. Furthermore, the target descriptive information can be appropriately stored in the RAG database 50 in accordance with prescribed rules without any operation by the administrator. This makes it possible to provide an LLM-based system that is easy for both employees and administrators to use. 【0112】 [7. Basic Computer Hardware Configuration] Figure 10 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 921. 【0113】 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. 【0114】 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). 【0115】 Auxiliary storage device 903 refers to a storage device for saving data and programs. Examples include flash memory, HDD (Hard Disk Drive), magneto-optical disk, CD-ROM, DVD-ROM, and semiconductor memory. 【0116】 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. 【0117】 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. 【0118】 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. 【0119】 [8. Basic Functional Configuration of a Computer] The functional configuration of the computer realized by the basic hardware configuration of computer 90 (Figure 10) is described below. The computer comprises at least one functional unit: a control unit, a memory unit, and a communication unit. 【0120】 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. 【0121】 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. 【0122】 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. 【0123】 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. 【0124】 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. 【0125】 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. 【0126】 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. 【0127】 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. 【0128】 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 embodiments. 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 embodiments 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, and the like. 【0129】 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, Java®, JavaScript, and TypeScript. 【0130】 Furthermore, the program code of the software that realizes the functions of the embodiment may be distributed via a network and stored on a storage means such as a computer's hard disk or memory, or on a storage medium such as a CD-RW or CD-R, and the computer's processor may read and execute the program code stored on the storage means or storage medium. 【0131】 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 (Central Processing Units), conventional circuits, and / or combinations thereof, programmed to realize the functions described herein. A processor is considered to be a circuit or processing circuitry, including transistors and other circuits. A processor may be a programmed processor that executes a program stored in memory. 【0132】 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. 【0133】 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. 【0134】 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. 【0135】 [10. Addendum] The details described in each of the above embodiments are noted below. 【0136】 <Note 1> A program for execution on a computer having a processor and memory, which stores information about multiple users and multiple pieces of information associated with each of the multiple user pieces of information for explaining predetermined content to the user, and the program causes the processor to execute the following steps: presenting the user with multiple choices required by the user and accepting the user's selection of one of the multiple choices; referring to the instruction sentences included in the accepted choices and information about the user who selected the choice, reading predetermined information from the multiple pieces of information; inputting the input instruction sentences and the read predetermined information into a large language model and causing the large language model to output a response to the input instruction sentences; and presenting the output response to the user who selected the choice. 【0137】 <Note 2> In the acceptance step, multiple options are output from a large-scale language model that outputs a response based on information about the user, and from a different large-scale language model (as described in Appendix 1). 【0138】 <Note 3> In the input step, the program described in Appendix 2 inputs user information and the instruction "Output multiple appropriate options for the user providing the input" into a large-scale language model different from the large-scale language model that outputs the response. 【0139】 <Note 4> A program as described in any of (Appendix 1) to (Appendix 3), wherein multiple users are multiple employees, information about multiple employees is accessible to multiple employees simultaneously, and in the receiving step, multiple options are presented to the employee based on the information about the accessed employee. 【0140】 <Note 5> A program described in any of (Appendix 1) to (Appendix 4) that causes the processor to further perform the steps of obtaining at least some of the information from a set of multiple pieces of information and storing the obtained information in accordance with a predetermined rule. 【0141】 <Note 6> An information processing device comprising a control unit and a storage unit, wherein the control unit executes all steps in any of the programs described in (Appendix 1) to (Appendix 5). 【0142】 <Note 7> A method to be executed on a computer having a processor and memory, wherein the processor executes all steps in any of the programs described in (Appendix 1) to (Appendix 5). 【0143】 <Note 8> A system comprising means for executing all steps in any of the programs described in (Appendix 1) to (Appendix 5). [Explanation of symbols] 【0144】 1... System 10…Terminal device 120... Communications Department 13…Input device 14…Output device 15…Memory 16…Storage 19… Processor 20... Server 22...Communication IF 23…Input / Output Interface 25…Memory 2 hours… storage 29… Processor 30…Generating AI system 40…Employee Database 50…Database for RAG

Claims

[Claim 1] A program to be executed on a computer having a processor and memory, Information about multiple users, and multiple pieces of information associated with each of the pieces of information about the multiple users, for explaining predetermined content to the user, are stored. The program is provided to the processor: The steps include presenting the user with a number of options required by the user and accepting the user's selection of any of the options, The steps include: reading predetermined information from the plurality of pieces of information by referring to the instruction text included in the received options and information about the user who selected the options; The steps include inputting the input instruction sentence and the predetermined information read out into a large-scale language model, and causing the large-scale language model to output a response to the input instruction sentence, The steps include presenting the outputted response to the user who selected the option, and A program that executes the command. [Claim 2] In the receiving step, the program according to claim 1, wherein the multiple options are output from a large-scale language model different from the large-scale language model that outputs the response, based on information about the user. [Claim 3] The program according to claim 2, wherein in the receiving step, the multiple options are input to a large-scale language model different from the large-scale language model that outputs the response, along with information about the user and the instruction "Output multiple appropriate options for the user who is inputting". [Claim 4] The aforementioned multiple users are multiple employees, The information relating to the aforementioned multiple employees is accessible to the aforementioned multiple employees simultaneously. The program according to claim 1, which, in the receiving step, presents the plurality of options corresponding to the employee based on the information of the employee that has accessed the information. [Claim 5] The program according to claim 1, which causes the processor to further perform the step of acquiring at least some of the aforementioned plurality of pieces of information and storing the acquired at least some of the information in accordance with a predetermined rule. [Claim 6] An information processing device comprising a control unit and a storage unit, wherein the control unit executes all steps in the program described in any one of claims 1 to 5. [Claim 7] A method to be performed on a computer comprising a processor and memory, wherein the processor performs all steps of a program according to any one of claims 1 to 5. [Claim 8] A system comprising means for executing all steps in the program described in any one of claims 1 to 5.

Images