Program, information processing device, manufacturing method, and information processing method

Abstract

To provide a program for obtaining a higher quality answer content relative to a question about a law received from a user, an information processing device, a manufacturing method, and an information processing method.SOLUTION: A program to be executed by a server, which is an information processing device constituted by a computer having a processor and a storage unit, causes the processor to execute: a question reception step of receiving a question from a user; a specifying step of specifying one or a plurality of predetermined consultation answer data from a plurality of consultation answer data composed of a question or an answer relative to past legal consultation based on the question received in the question reception step; and a generation step of generating a prompt, which is an input sentence relative to an interaction response AI, based on the question received in the question reception step and the predetermined consultation answer data specified in the specification step.SELECTED DRAWING: Figure 9

Description

[Technical Field]

[0001] The present disclosure relates to a program, an information processing device, a manufacturing method, and an information processing method. [Background technology]

[0002] 2. Description of the Related Art Web services that provide legal cases on websites are known. Patent Document 1 discloses a legal search system that allows users to search for desired articles and court cases without knowing legal search keywords, allows users to easily and quickly search by using combinations of verbs and objects, and allows users to obtain desired results by sequentially selecting from appropriate options. [Prior art documents] [Patent documents]

[0003] [Patent Document 1] Japanese Patent Application Laid-Open No. 2002-140339 Summary of the Invention [Problem to be solved by the invention]

[0004] There is a problem in that it is not possible to obtain high-quality answers to legal questions received from users. Therefore, the present disclosure has been made to solve the above problem, and its purpose is to provide a technology for obtaining higher quality answers to legal questions received from users. [Means for solving the problem]

[0005] A program to be executed by a computer having a processor and a memory unit, the program executing the following steps: a question receiving step in which the processor receives a question from a user; a specification step in which, based on the question received in the question receiving step, it identifies one or more predetermined consultation answer data from a plurality of consultation answer data consisting of questions or answers to past legal consultations; and a generation step in which, based on the question received in the question receiving step and the predetermined consultation answer data identified in the specification step, it generates a prompt, which is an input sentence for a dialogue response AI. [Effects of the Invention]

[0006] According to the present disclosure, higher quality answers can be obtained in response to legal questions received from users. [Brief explanation of the drawings]

[0007] [Figure 1] FIG. 2 is a block diagram showing the functional configuration of the system 1. [Figure 2] FIG. 2 is a block diagram showing the functional configuration of the server 10. [Figure 3] FIG. 2 is a block diagram showing the functional configuration of a user terminal 20. [Figure 4] FIG. 10 is a diagram showing the data structure of a user table 1012. [Figure 5] FIG. 10 is a diagram showing the data structure of a question table 1013. [Figure 6] FIG. 10 is a diagram showing the data structure of a response table 1014. [Figure 7] FIG. 10 is a diagram showing the data structure of a question log table 1021. [Figure 8] FIG. 10 is a diagram showing the data structure of a reply log table 1022. [Figure 9] 10 is a flowchart showing the operation of a question and answer process. [Figure 10] 10 is a screen example showing the operation of the question and answer process. [Figure 11] FIG. 2 is a block diagram showing the basic hardware configuration of a computer 90. DETAILED DESCRIPTION OF THE INVENTION

[0008] Hereinafter, embodiments of the present disclosure will be described with reference to the drawings. In all drawings describing the embodiments, common components are designated by the same reference numerals, and repeated description will be omitted. Note that the following embodiments do not unduly limit the content of the present disclosure described in the claims. Furthermore, not all components shown in the embodiments are necessarily essential components of the present disclosure. Furthermore, each drawing is a schematic diagram and is not necessarily a precise illustration.

[0009] <System 1 Configuration> System 1 in this disclosure is an information processing system that generates answers to legal questions. The system 1 includes a server 10, a user terminal 20, and an information processing device of an LLM system 50, which are connected via a network N. FIG. 1 is a block diagram showing the functional configuration of the system 1. As shown in FIG. FIG. 2 is a block diagram showing the functional configuration of the server 10. As shown in FIG. FIG. 3 is a block diagram showing the functional configuration of the user terminal 20. As shown in FIG.

[0010] Each information processing device is configured by a computer equipped with an arithmetic unit and a storage device. The basic hardware configuration of the computer and the basic functional configuration of the computer realized by this hardware configuration will be described later. For each of the server 10, the user terminal 20, and the LLM system 50, descriptions that overlap with the basic hardware configuration and basic functional configuration of the computer will be omitted.

[0011] <Server 10 configuration> The server 10 is an information processing device that provides an information processing service for creating answers to legal questions. The server 10 includes a storage unit 101 and a control unit 104 .

[0012] <Configuration of the storage unit 101 of the server 10> The storage unit 101 of the server 10 includes an application program 1011 , a user table 1012 , a question table 1013 , an answer table 1014 , a question log table 1021 , and an answer log table 1022 .

[0013] The application program 1011 is a program for causing the control unit 104 of the server 10 to function as each functional unit. Application programs 1011 include applications such as a web browser application.

[0014] User table 1012 is a table that stores and manages information about member users (hereinafter, "users") who use the service. When a user registers to use the service, the user's information is stored in a new record in user table 1012. This allows the user to use the service according to the present disclosure. The user table 1012 is a table having columns of user IDs and user names, with the user ID as the primary key. FIG. 4 is a diagram showing the data structure of the user table 1012.

[0015] The user ID is an item that stores user identification information for identifying a user. The user identification information is an item that is set with a unique value for each user. The user name is an item for storing the name of the user. The user name may be set to any character string such as a nickname instead of a name.

[0016] The question table 1013 is a table for storing and managing information relating to questions (question information). The question table 1013 is a table having a question ID as a primary key, and columns of a question ID, a parent question ID, a user ID, question data, and a date and time. FIG. 5 is a diagram showing the data structure of the question table 1013.

[0017] The question ID is an item that stores question identification information for identifying a question. The question identification information is an item that has a unique value set for each piece of question information. The parent question ID is an item for storing question identification information for identifying a parent question. The user ID is an item that stores user identification information for identifying a user. Note that the user does not need to be registered in advance with the information processing service according to the present disclosure. In this case, the user ID does not need to be stored, and any information for identifying the user, such as the user's IP address or cookie information, may be stored. The question data is an item for storing a question about the law received from the user. The date and time is an item for storing the date and time when a new record is stored or updated in the question table.

[0018] The answer table 1014 is a table for storing and managing information about answers (answer information). The answer table 1014 is a table having columns for question ID, prompt, answer data, and date and time. FIG. 6 is a diagram showing the data structure of the answer table 1014.

[0019] The question ID is an item for storing question identification information for identifying a question. The prompt is an item for storing an input sentence to be sent to the LLM system 50, which will be described later. The response data is an item for storing a response to an input statement sent to the LLM system 50. The date and time is an item for storing the date and time when a new record is stored or updated in the answer table.

[0020] Question log table 1021 is a table for storing and managing information (question log information) about questions asked in past legal consultations. Specifically, the content of questions that users have asked in the past on web services that allow users to ask legal questions is stored as question log data. The question log table 1021 is a table having a question log ID as a primary key, and columns of a question log ID, a user ID, question log data, and a date and time. FIG. 7 is a diagram showing the data structure of the question log table 1021. As shown in FIG.

[0021] The question log ID is an item that stores question log identification information for identifying a question log. The question log identification information is an item that is assigned a unique value for each piece of question log information. The user ID is an item for storing user identification information for identifying a user. The question log data is an item for storing the content of questions asked by users. Specifically, the question log data stores text data of legal questions, image data, other attachments, and the like. The date and time is an item for storing the date and time when a new record is stored or updated in the question log table.

[0022] Answer log table 1022 is a table for storing and managing information (answer log information) related to answers to past legal consultations. Specifically, the answers given by experts to the question log data stored in question log table 1021 are stored as answer log data. The answer log table 1022 is a table having columns for question log ID, user ID, answer log data, evaluation log data, and date and time. FIG. 8 is a diagram showing the data structure of the response log table 1022. As shown in FIG.

[0023] The question log ID is an item for storing question log identification information for identifying a question log. The user ID is an item for storing user identification information for identifying a user. The answer log data is an item for storing answers by experts. Specifically, the answer log data stores text data, image data, other attachments, etc. of answers by experts to legal questions. The evaluation log data is an item for storing evaluations of the answer contents. The questioner can select the answer that best suits the question from among the multiple answers (answer log data) given by multiple experts as the "best answer." The evaluation log data includes information indicating whether the answer log data is the best answer. Experts can evaluate (react) to the answers (answer log data) provided by other experts by giving them a thumbs up, rating, etc. The evaluation log data includes information indicating which expert reacted to the answer log data and what kind of reaction they made. Note that reactions do not have to be made only by experts, and the system may be configured to store reactions made by the questioner and general users other than the questioner. The date and time is an item for storing the date and time when a new record is stored or updated in the reply log table.

[0024] <Configuration of the control unit 104 of the server 10> The control unit 104 of the server 10 includes a user registration control unit 1041 and a question and answer unit 1042. The control unit 104 executes an application program 1011 stored in the storage unit 101, thereby realizing each functional unit.

[0025] The user registration control unit 1041 performs processing to store information about users who wish to use the service according to the present disclosure in the user table 1012. The information stored in the user table 1012 is generated when a user opens a web page operated by a service provider from any information processing terminal, enters information into a predetermined input form, and transmits the information to the server 10. The user registration control unit 1041 stores the received information in a new record in the user table 1012, completing the user registration. This allows the user stored in the user table 1012 to use the service. Before the user registration control unit 1041 registers the user information in the user table 1012, the service provider may conduct a predetermined examination to restrict whether or not the user is permitted to use the service. The user ID may be any character string or number that can identify the user, any character string or number desired by the user, or may be automatically set by the user registration control unit 1041.

[0026] The question answering unit 1042 executes the question answering process, the details of which will be described later.

[0027] <Configuration of User Terminal 20> The user terminal 20 is an information processing device operated by a user who uses a service. The user terminal 20 may be, for example, a mobile terminal such as a smartphone or tablet, a stationary personal computer (PC) or a laptop PC, or a wearable terminal such as a head mounted display (HMD) or a wristwatch terminal. The user terminal 20 includes a storage unit 201 , a control unit 204 , an input device 206 , and an output device 208 .

[0028] <Configuration of the storage unit 201 of the user terminal 20> The storage unit 201 of the user terminal 20 includes a user ID 2011 and an application program 2012 .

[0029] The user ID 2011 is the user's account ID. The user transmits the user ID 2011 from the user terminal 20 to the server 10. The server 10 identifies the user based on the user ID 2011 and provides the user with the service according to the present disclosure. The user ID 2011 includes information such as a session ID temporarily assigned by the server 10 to identify the user using the user terminal 20.

[0030] The application program 2012 may be stored in advance in the storage unit 201, or may be configured to be downloaded from a web server or the like operated by a service provider via a communication IF. The application programs 2012 include applications such as a web browser application. The application program 2012 includes an interpreter-type programming language such as JavaScript (registered trademark) that runs on a web browser application stored in the user terminal 20.

[0031] <Configuration of the control unit 204 of the user terminal 20> The control unit 204 of the user terminal 20 includes an input control unit 2041 and an output control unit 2042. The control unit 204 realizes each functional unit by executing the application program 2012 stored in the storage unit 201.

[0032] <Configuration of the input device 206 of the user terminal 20> The input device 206 of the user terminal 20 includes a camera 2061, a microphone 2062, a position information sensor 2063, a motion sensor 2064, and a touch device 2065.

[0033] <Configuration of the output device 208 of the user terminal 20> The output device 208 of the user terminal 20 includes a display 2081 and a speaker 2082.

[0034] <Configuration of the LLM system 50> The LLM system 50 is an information processing device that provides an interactive response AI service. Specifically, the LLM system 50 has an LLM (Large Language Model) internally and can output an output sentence (answer sentence) for an input sentence called a prompt. Well-known LLMs include ChatGPT and BERT. The LLM system 50 may be operated by an information processing service provider according to the present disclosure, or may be used via an API (Application Programming Interface) provided by an external LLM system 50 operated by another operator.

[0035] <Operation of the system 1> Hereinafter, each process of the system 1 will be described. FIG. 9 is a flowchart showing the operation of the question-and-answer process. FIG. 10 is an example of a screen showing the operation of the question and answer process.

[0036] <Question and answer processing> The question answering process is a process of presenting an answer from an artificial intelligence model to a user in response to a question received from the user.

[0037] <Question and answer processing overview> The question and answer process is a series of processes that accept questions from users, search multiple consultation and answer data consisting of questions or answers from past legal consultations, generate prompts that are input sentences for the dialogue response AI, output the prompts to the dialogue response AI, receive the answer content, present the answer content, and create a legal consultation card for consulting an expert in response to input operations from the user.

[0038] <Question and answer processing details> The question and answer process will be described in detail below.

[0039] <Question Reception (First Embodiment)> In step S101, the question and answer section 1042 of the server 10 executes a question receiving step of receiving a question from a user. Specifically, the user operates the input device 206 of the user terminal 20 to execute a browser application or the like, and opens the question and answer page D1 by inputting the URL of a web page (question and answer page) for executing the question and answer process. The control unit 204 of the user terminal 20 sends a request including the user ID 2011 for opening the question and answer page to the server 10.

[0040] When the server 10 receives the request, it generates a question and answer page and transmits it to the user terminal 20. The control unit 204 of the user terminal 20 displays the question and answer page on the display 2081 of the user terminal 20 and presents it. 10 is an example of a question and answer screen in the question and answer process. A question and answer page D1 is displayed on the display 2081 of the user terminal 20. The question and answer page D1 includes a question input field D101, a send button D102, questions D111 and D112 entered by the user, a response D121 by the dialogue response AI in response to a prompt generated based on the question, related articles D131, D132, and D133, icons D141, D142, and D143 indicating experts in the related articles, and a create button D151 for creating a legal advice card.

[0041] The user inputs a question into the question input field D101 by operating the input device 206 of the user terminal 20. In the present disclosure, the user inputs a legal question such as "Please tell me the going rate for divorce compensation." The user presses the send button D102 by operating the input device 206 of the user terminal 20. The control unit 204 of the user terminal 20 transmits the user ID 2011 and the character string entered in the question input field D101 (the question from the user) to the server 10.

[0042] The question and answering unit 1042 of the server 10 receives and accepts a question from a user. The question and answering unit 1042 of the server 10 stores the received user ID 2011 and the question from the user in the user ID and question data items of the question table 1013, respectively. The question ID stores question identification information that identifies the stored question information.

[0043] <Question Reception (Second Embodiment)> In step S101, a question receiving step is executed to receive a plurality of questions from the user. Specifically, the question input field D101 of the question and answer page D1 in the present disclosure is configured to be able to accept multiple questions from the user. For example, the user operates the input device 206 of the user terminal 20 to input and send a question D111 into the question input field D101. Thereafter, in step S107 of the question and answer process, the control unit 204 of the user terminal 20 presents an answer D121 to the user. The user can operate the input device 206 of the user terminal 20 to input and send a question D112 into the question input field D101. The control unit 204 of the user terminal 20 stores the user ID 2011, the question from the user, and the question identification information associated with the question D111 in the items of user ID, question data, and parent question ID in the question table 1013. As a result, the question D111 and the questions D112, D113, etc., asked subsequently are stored in association with each other. As a result, the questions D111 and D112 entered by the user and the answer D121 by the dialogue response AI in response to the prompt generated based on the questions are displayed on the display 2081 of the user terminal 20, as shown in the question and answer page D1 in Figure 10.

[0044] <Question Log Search (First Embodiment)> In step S102, the question answering unit 1042 of the server 10 executes an identification step of identifying one or more predetermined consultation answer data from a plurality of consultation answer data consisting of questions or answers to past legal consultations, based on the question received in the question receiving step. Specifically, the question answering unit 1042 of the server 10 searches the question log data items in the question log table 1021 based on the question text from the user received in step S101, and acquires multiple pieces of question log information including the question log ID, the question log data, and the date and time items. In addition, the question answering unit 1042 of the server 10 may search the answer log data items in the answer log table 1022 based on the question text from the user received in step S101, and obtain multiple answer log information items including the question ID, answer log data, evaluation log data, and date and time.

[0045] The search will be described as an example of a method for performing a full-text search of question log data or answer log data based on a question sentence. The full-text search is performed based on the degree of match between words obtained by morphological analysis of the question sentence and words obtained by morphological analysis of sentences included in the question log data or answer log data. Search algorithms include TF-IDF, LexRank, and methods that calculate the cosine similarity between the question sentence and the question log data or answer log data. Any search method can be used. The multiple pieces of question log information and answer log information obtained through the search are associated with numerical indices such as similarity and relevance (hereinafter referred to as similarity) calculated based on the similarity with the question text. The question and answer unit 1042 of the server 10 acquires and identifies, as search results, a predetermined number (for example, five) of question log information and answer log information in descending order of similarity, or multiple pieces of question log information and answer log information whose similarity is equal to or greater than a predetermined value.

[0046] If the similarity between the question message from the user and the question log information and answer log information obtained by the search is equal to or less than a predetermined value (if no question log information or answer log information related to the question message from the user is found), the process may be terminated without proceeding to step S103, assuming that no search results were obtained. In this case, the user may be notified with a message such as "We are unable to fully answer your question." This makes it possible to avoid generating an inappropriate answer based on the generated prompt when the content of the consultation answer data searched based on the question text has low relevance to the question text.

[0047] In step S102, the specifying step executes a step of specifying predetermined consultation answer data by searching for the question in a plurality of consultation answer data based on the question received in the question receiving step. Specifically, it is preferable that the question answering unit 1042 of the server 10 searches the question log data in the question log table 1021, rather than the answer log data stored in the answer log table 1022, based on the question text from the user received in step S101. This is because the question log information, not the answer log information, is more relevant to the question from the user. By including past consultation and answer data in the prompts input to the dialogue response AI, it is possible to obtain higher quality answers to questions received from users.

[0048] <Question Log Search (Second Embodiment)> In step S102, the specifying step executes a step of specifying predetermined consultation answer data based on a question obtained by combining a plurality of questions received in the question receiving step. Specifically, a case where multiple questions are received in step S101, as in the case of the second embodiment, will be described. The question answering unit 1042 of the server 10 searches the question ID item and the parent question ID item in the question table 1013 based on the received parent question ID, and acquires question information including data on questions asked in the past. The question and answer unit 1042 of the server 10 sorts the acquired question data in order of date and time, and combines each question sentence to create a question sentence. For example, the question data (first question data) of the first question D111 is "Please tell me the going rate for alimony." The question data (second question data) of the second question D112 is "Whether my spouse is having an affair or not. My husband is having an affair. What is my relationship with the other person and how do I treat them? It seems that they meet about three days a week." In this case, the question answering unit 1042 of the server 10 combines the first question data and the second question data to generate the question "Please tell me the going rate for alimony. Whether my spouse is having an affair or not. My husband is having an affair. What is my relationship with the other person and how do I treat them? It seems that they meet about three days a week." The question and answer unit 1042 of the server 10 searches the question log table 1021 and the answer log table 1022 based on the generated question sentence, similar to the question log search (first embodiment). This makes it possible to obtain higher quality answers based on questions that combine multiple questions.

[0049] The question log search (first embodiment) and the question log search (second embodiment) may be configured to search only question log information and answer log information within a certain period (within two years) based on the date and time items of the question log information and answer log information. This is because, for example, due to legal amendments or the like, old question log information and answer log information may not be useful at the time the user asks a question. It is preferable to search for newer question log information and answer log information whenever possible.

[0050] <Response log identification> In step S103, the question and answering unit 1042 of the server 10 identifies answer log information associated with the question log information searched for in step S102. Specifically, the question and answering unit 1042 of the server 10 searches the question log ID field in the answer log table 1022 based on the question log ID included in the question log information searched for in step S102, and acquires multiple pieces of answer log information including answer log data, user ID, evaluation log data, and date and time. In general, multiple pieces of answer log information are associated with one piece of question log information and one piece of answer log information. For example, multiple answers (answer logs) are provided by multiple experts in response to one question (question log) from a user. The question and answer unit 1042 of the server 10 may identify all or some of the answer log information associated with the question log searched in step S102. Hereinafter, several embodiments will be described as a method for identifying some of the answer log information.

[0051] <Response Log Identification (First Embodiment)> In step S103, the identifying step includes a step of excluding one or more answers from the predetermined consultation answer data based on information indicating that the answer is appropriate to the question, among one or more answers included in the plurality of consultation answer data. Specifically, the question answering unit 1042 of the server 10 refers to the evaluation log data contained in all answer log information associated with the question log searched in step S102, identifies some answer log information that stores information indicating that the answer is the "best answer" selected by the asker as the answer most appropriate to the question, and identifies answer log information by excluding other answer log information. This allows the AI ​​to obtain higher quality answers to questions received from users by excluding low-quality answers from the actual consultation response data that has been given in the past in response to prompts entered into the AI. The question answering unit 1042 of the server 10 does not necessarily need to exclude all answer log information that is not the "best answer," but may be configured to exclude at least one answer log information. For example, this embodiment may be applied in combination with other embodiments of specifying answer logs.

[0052] <Response Log Identification (Second Embodiment)> In step S103, the identifying step includes a step of excluding one or more answers from the predetermined consultation answer data based on evaluation information stored in association with the answer, among one or more answers included in the plurality of consultation answer data. The evaluation information is information based on reactions given by predetermined experts. Specifically, the question answering unit 1042 of the server 10 refers to the evaluation log data included in all the answer log information associated with the question log searched in step S102, identifies some of the answer log information to which evaluations (reactions) such as "like" or "rating" have been assigned, and identifies the answer log information by excluding other answer log information. Furthermore, the question answering unit 1042 of the server 10 may identify some of the answer log information based on the number of reactions, the quality of the reactions, etc. It may also be configured to consider only reactions given by designated experts such as lawyers, etc. This allows the AI ​​to obtain higher quality answers to questions received from users by excluding low-quality answers based on highly accurate evaluation information given by experts from the actual consultation response data in the past in response to prompts input to the AI. The question answering unit 1042 of the server 10 does not necessarily need to exclude all answer log information to which no reaction is added, but may be configured to exclude at least one answer log information. For example, this embodiment may be applied in combination with other embodiments of specifying answer logs.

[0053] <Response Log Identification (Third Embodiment)> In step S103, the identifying step includes a step of excluding one or more answers from the predetermined consultation answer data based on the answerer who provided the answer, among one or more answers included in the plurality of consultation answer data. Specifically, the question answering unit 1042 of the server 10 refers to the user IDs contained in all answer log information associated with the question logs searched in step S102, identifies some of the answer log information based on the information of the answerer identified based on the user ID, and identifies the answer log information by excluding other answer log information. For example, information that can be used to identify answer log information includes the number of years of experience of the expert who answered, the registration number for qualifications, etc., the number of reactions such as likes to the expert, the number of answer log information stored in answer log table 1022 (the number of past answers by the expert), the number of reactions such as likes to answer log information to which the expert has previously answered, information regarding the expert's work history as a former judge, former prosecutor, etc., and information such as field of expertise. For example, the question and answer unit 1042 of the server 10 may calculate an evaluation index value for the expert based on the above information, identify some answer log information based on the evaluation index value, and identify answer log information by excluding other answer log information. The evaluation index value may be obtained as output data by applying any machine learning model, deep learning model, artificial intelligence model, etc. to the above information as input data. Furthermore, answer log information may be identified by identifying some answer log information and excluding other answer log information based on the degree of association between the genre and category identified based on the question received in step S101 and the specialty field of the expert. For example, if the question is about "divorce," answer log information by experts whose specialty field is "divorce" may be prioritized, and answer log information by experts in other specialty fields may be excluded. The genre and category of a question may be obtained as output data by applying any machine learning model, deep learning model, artificial intelligence model, etc. to the question as input data. For example, the model can be obtained by learning the areas of expertise of respondents related to answer log data for past question log data as training data.

[0054] <Prompt Generation> In step S104, the question answering unit 1042 of the server 10 executes a generating step of generating a prompt, which is an input sentence to the dialogue response AI, based on the question received in the question receiving step and the predetermined consultation answer data identified in the identifying step. The generating step executes a step of generating a prompt including the predetermined consultation answer data consisting of the question searched in the identifying step and the answer to the question.

[0055] Specifically, the question answering unit 1042 of the server 10 generates a prompt, which is an input sentence for the dialogue response AI, by combining the question sentence received in step S101, the question log information (question log data) searched in step S102, and the answer log information (answer log data) identified in step S103. For example, suppose the question is "What is the going rate for divorce compensation?" and the question log data and answer log data are "Q1, A11, A12," "Q2, A2," and "Q3, A3, A31," respectively. Note that Q1, Q2, and Q3 are character strings related to the question log data. A11, A12, A2, and A31 are character strings related to the answer log data. In this case, the question and answer section 1042 of the server 10 generates the prompt as follows: " Please write your answer to the question based on the reference case below. #Question Please tell me the average amount of divorce compensation #Reference example Q:Q1 A: A11, A12 Q:Q2 A:A2 Q:Q3 A: A3, A31 "

[0056] In the present disclosure, the prompt includes both the question log information and the answer log information, but it may also be configured to include only either the question log information or the answer log information.

[0057] In step S104, the generating step does not generate a prompt if the degree of association between the question received in the question receiving step and the predetermined consultation answer data identified in the identifying step is equal to or less than a predetermined value. Specifically, if the question log search in step S102 determines that the question log information and answer log information obtained in steps S102 and S103 are insufficient for the user's question, or that the degree of relevance with the question is lower than a predetermined value, no prompt is generated and the process does not proceed to step S105. For example, the degree of relevance may be determined as the cosine similarity between the user's question and the question log data or answer log data. This makes it possible to avoid generating an inappropriate answer based on the generated prompt when the content of the consultation answer data searched based on the question text has low relevance to the question text.

[0058] In step S105, the question and answer unit 1042 of the server 10 executes an output step of outputting the prompt generated in the generation step to an LLM (Large Scale Language Model). Specifically, the question answering unit 1042 of the server 10 sends a request including the prompt generated in step S104 to an API endpoint related to the dialogue response AI service provided by the LLM system 50.

[0059] In step S106, the question answering unit 1042 of the server 10 executes a receiving step of receiving, from the LLM, the content of the answer to the prompt output in the output step. Specifically, the LLM system 50 transmits a response to the received request to the server 10. The response includes a character string that is an answer to the prompt included in the request. The question answering unit 1042 of the server 10 receives and accepts the answer to the prompt.

[0060] <Answer Presentation (First Embodiment)> In step S107, the question and answer section 1042 of the server 10 executes a presentation step of presenting the answer content received in the reception step. Specifically, the question and answer unit 1042 of the server 10 transmits the received answer sentence to the user terminal 20. The control unit 204 of the user terminal 20 displays the received answer sentence D121 on the question and answer page D1 and presents it to the user.

[0061] In step S107, a presentation step is executed in which the answer content is presented in association with at least a part of the predetermined consultation answer data identified in the identification step. Specifically, the question and answer unit 1042 of the server 10 transmits the related articles D131, D132, and D133 corresponding to the question log data identified in step S102 to the user terminal 20. The control unit 204 of the user terminal 20 associates the related articles D131, D132, and D133 with the answer sentence D121, displays them on the question and answer page D1, and presents them to the user. Note that the character strings included in the related articles D131, D132, and D133 may be part of the question log data, a summary of the character strings included in the answer log data associated with the question log data, a title, or other heading. This allows users to view the original consultation response data along with the response content from LLM, allowing users to refer to more comprehensive information regarding their questions.

[0062] In step S107, the presenting step includes a step of presenting at least a part of the predetermined consultation answer data in association with information about the expert who provided the answer included in the predetermined consultation answer data. Specifically, the question answering unit 1042 of the server 10, for each piece of question log data identified in step S102, references the user table 1012 based on the user ID included in the answer log information identified in step S103, and transmits information such as the name and photo of the answerer identified by the user table 1012 to the user terminal 20. The control unit 204 of the user terminal 20 generates icons D141, D142, and D143 indicating experts on related articles based on the information such as the answerer's name and photo, associates the icons with the related articles D131, D132, and D133 related to the question log information answered by the respective answerers, and displays them on the question answering page D1 to present to the user. The user operates the input device 206 of the user terminal 20 to select icons D141, D142, and D143, which causes an input form such as a message form to be displayed, allowing the user to contact the respective experts through the input form. This allows users to view the response from LLM along with information about the expert who provided the answer, which serves as the source for generating the response, by opening the link. Users can easily find experts who can respond to their questions.

[0063] <Answer Presentation (Second Embodiment)> In step S107, a presentation step is executed in which a link for referencing predetermined consultation answer data is presented in association with at least a part of the predetermined consultation answer data. Specifically, the question and answer unit 1042 of the server 10 sets links to the related articles D131, D132, and D133 to individual article pages related to the question log data that is the source of the related articles. The user can select related articles D131, D132, and D133 by operating the input device 206 of the user terminal 20, and thereby open the individual article pages related to the question log data associated with each of the related articles. For example, an individual article page includes information about the questioner (user name, name, photo, etc., contact information such as email address) identified by referencing user table 1012, etc. based on the user ID included in the question log information, question log data which is the question text, and information on the date and time of the question. An individual article page includes one or more pieces of answer log information for question log information identified by referencing the question log ID in answer log table 1022, based on the question log ID included in the question log information. An individual article page includes information about the answerer (user name, name, photo, etc., contact information such as email address) identified by referencing user table 1012, etc. based on the user ID included in the answer log information, answer log data which is the answer text, evaluation log data such as the best answer and reactions, and information on the date and time of the answer. This allows users to refer to the consultation and response data that was used as the source for generating the response content, along with the response content from LLM, allowing users to obtain more comprehensive information regarding their questions.

[0064] <Answer Presentation (Third Embodiment)> In step S107, the presentation step executes a step of not presenting a link for referring to predetermined consultation answer data. The question and answer unit 1042 of the server 10 may not set links to the individual article pages related to the question log data that is the source of the related articles D131, D132, and D133. This makes it possible to prevent users from referencing the consultation answer data that was the source when generating the answer content along with the answer content from LLM. It is possible to present only the answer content from LLM and restrict reference to the consultation answer data that was the source. In some cases, it may be preferable not to set a link to the source in this way depending on the content of the service provided. This makes it possible to limit the information provided to users.

[0065] <Card creation process> In step S108, the question and answer unit 1042 of the server 10 executes a card creation step to create a legal consultation card for when consulting with an expert, based on the question received in the question receiving step and the answer content received in the receiving step. Specifically, the user presses the create button D151 by operating the input device 206 of the user terminal 20. The control unit 204 of the user terminal 20 transmits a request to create a legal advice card to the server 10. Upon receiving the request, the question and answer unit 1042 of the server 10 creates a text (legal advice card) related to legal advice when consulting a legal expert such as an external lawyer, based on the questions D111, D112, and the answer D121. For example, the question and answer unit 1042 of the server 10 combines the character strings of the questions D111 and D112 to create a text related to the content of the legal advice, and applies this to a legal advice card template document (not shown) to create the legal advice card. The question and answer unit 1042 of the server 10 transmits the created legal advice card to the user terminal 20. The control unit 204 of the user terminal 20 displays and presents the received legal advice card on the display 2081 of the user terminal 20. In addition, the user may include information about the expert to be consulted (such as name, photo, contact information such as email address, etc.) in the legal consultation card by selecting icons D141, D142, and D143 by operating the input device 206 of the user terminal 20. This makes it easy to create a legal consultation card that is required when consulting with a specialist.

[0066] In step S108, the question and answer unit 1042 of the server 10 executes an inquiry step of inquiring of a plurality of experts as to whether they are available to handle the consultation, based on the legal consultation card created in the card creation step. Specifically, when transmitting a legal advice card to the user terminal 20, the question answering unit 1042 of the server 10 transmits information on one or more experts associated with the legal advice card to the user terminal 20. For example, the experts are preferably experts associated with the related articles D131, D132, and D133, and in particular experts who have actually answered similar questions in the past in the answer log information. The control unit 204 of the user terminal 20 presents the received icons representing one or more experts to the user along with the legal advice card. The user operates the input device 206 of the user terminal 20 to select the presented icon, thereby transmitting a request to the server 10 to provide legal advice to the expert, including the legal advice card. Based on the received request, the question and answer unit 1042 of the server 10 inquires of the selected expert whether or not they are able to handle the legal advice, including the legal advice card. This allows the user to easily find an expert who can respond to the user's question.

[0067] In step S108, a card creation step is executed in which a card is created based on a sentence obtained by outputting the question accepted in the question acceptance step and the answer received in the reception step to the LLM. The legal advice card may be created based on the answer content obtained by sending the questions D111, D112, and answer D121 together with a predetermined prompt to the LLM system 50. For example, the question and answer unit 1042 of the server 10 generates a prompt such as the following and sends it to the LLM system 50. " Please prepare a document to consult with a legal expert based on the questions, answers, and reference cases below. #Question 1 Please tell me the average amount of divorce compensation #Answer 1 Well, the market price will vary depending on the specific circumstances and background, so you will need the following information: #Reference example Q:Q1 A: A11, A12 Q:Q2 A:A2 Q:Q3 A: A3, A31 #Question 2 Whether your spouse is having an affair: My husband is having an affair. Relationship with his lover and how he treats her: They seem to meet about three days a week. #Answer text 2 ... "

[0068] The question and answer unit 1042 of the server 10 generates a legal advice card based on the answer content (legal advice document) received from the LLM system 50, and transmits it to the user terminal 20 to present it to the user.

[0069] <Basic computer hardware configuration> 11 is a block diagram showing the basic hardware configuration of a computer 90. The computer 90 includes at least a processor 901, a main memory device 902, an auxiliary memory device 903, and a communication IF 991 (interface), which are electrically connected to one another by a communication bus 921.

[0070] The processor 901 is hardware for executing an instruction set written in a program, and is composed of an arithmetic unit, registers, peripheral circuits, and the like.

[0071] The main memory device 902 is used to temporarily store programs, data to be processed by the programs, etc. For example, it is a volatile memory such as a DRAM (Dynamic Random Access Memory).

[0072] The auxiliary storage device 903 is a storage device for saving data and programs, such as a flash memory, a hard disk drive (HDD), a magneto-optical disk, a CD-ROM, a DVD-ROM, or a semiconductor memory.

[0073] The communication IF 991 is an interface for inputting and outputting signals for communicating with other computers via a network using wired or wireless communication standards. The network is composed of the Internet, a LAN, various mobile communication systems constructed by wireless base stations, etc. For example, the network includes 3G, 4G, and 5G mobile communication systems, LTE (Long Term Evolution), and wireless networks (e.g., Wi-Fi (registered trademark)) that can connect to the Internet via a predetermined access point. In the case of a wireless connection, communication protocols include, for example, Z-Wave (registered trademark), ZigBee (registered trademark), and Bluetooth (registered trademark). In the case of a wired connection, the network also includes a direct connection using a USB (Universal Serial Bus) cable, etc.

[0074] It should be noted that the computer 90 can be virtually realized by distributing all or part of each hardware configuration across multiple computers 90 and interconnecting them via a network. In this way, the computer 90 is a concept that includes not only a computer 90 housed in a single housing or case, but also a virtualized computer system.

[0075] <Basic functional configuration of computer 90> The following describes the functional configuration of a computer realized by the basic hardware configuration (FIG. 11) of the computer 90. The computer includes at least the functional units of a control unit, a storage unit, and a communication unit.

[0076] The functional units of the computer 90 can also be realized by distributing all or part of the functional units among multiple computers 90 interconnected via a network. The computer 90 is a concept that includes not only a single computer 90 but also a virtualized computer system.

[0077] The control unit is realized by the processor 901 reading out various programs stored in the auxiliary storage device 903, expanding them in the main storage device 902, and executing processing in accordance with the programs. The control unit can realize functional units that perform various types of information processing depending on the type of program. In this way, the computer is realized as an information processing device that performs information processing.

[0078] The storage unit is realized by a main storage device 902 and an auxiliary storage device 903. The storage unit stores data, various programs, and various databases. Furthermore, the processor 901 can allocate a storage area corresponding to the storage unit in the main storage device 902 or the auxiliary storage device 903 in accordance with the programs. Furthermore, the control unit can cause the processor 901 to execute processes for adding, updating, and deleting data stored in the storage unit in accordance with the various programs.

[0079] A database refers to a relational database, which manages data sets called masters and tables in a tabular format structurally defined by rows and columns, by relating them to each other. In a database, a table is called a table, a master, a column in a table is called a column, and a row in a table is called a record. In a relational database, relationships between tables and masters can be set and associated. Typically, each table and each master has a column set as a primary key to uniquely identify a record, but setting a primary key to a column is not essential. The control unit can cause the processor 901 to add, delete, or update records in specific tables and masters stored in the storage unit according to various programs. Furthermore, by storing data, various programs, and various databases in the storage unit, it can be considered that the information processing device and information processing system according to the present disclosure have been manufactured.

[0080] Note that the databases and masters in this disclosure may include any data structure in which information is structurally defined (such as a list, dictionary, associative array, or object). The data structure also includes data that can be considered as a data structure by combining data with functions, classes, methods, etc. written in any programming language.

[0081] The communication unit is realized by the communication IF 991. The communication unit realizes a function of communicating with other computers 90 via a network. The communication unit can receive information transmitted from other computers 90 and input the information to the control unit. The control unit can cause the processor 901 to execute information processing on the received information in accordance with various programs. In addition, the communication unit can transmit information output from the control unit to other computers 90.

[0082] <Additional Notes> The matters described in the above embodiments will be supplemented below.

[0083] (Appendix 1) A program to be executed by a computer having a processor and a memory unit, the program executing the following steps: a question receiving step (101) in which the processor receives a question from a user; a specification step (S102, S103) in which, based on the question received in the question receiving step, one or more predetermined consultation answer data from a plurality of consultation answer data consisting of questions or answers to past legal consultations; and a generation step (S104) in which, based on the question received in the question receiving step and the predetermined consultation answer data specified in the specification step, a prompt is generated, which is an input sentence for a dialogue response AI. This allows the prompts input to the dialogue response AI to include data on actual consultation responses given in the past, making it possible to obtain higher quality answers to questions received from users.

[0084] (Appendix 2) The program according to appendix 1, wherein the identifying step (S102) is a step of identifying predetermined answer-to-consultation data by searching for a question in a plurality of pieces of answer-to-consultation data based on the question accepted in the question accepting step. This allows the prompts input to the dialogue response AI to include data on actual consultation responses given in the past, making it possible to obtain higher quality answers to questions received from users.

[0085] (Appendix 3) The program according to claim 1, wherein the generating step (S104) does not generate a prompt if the degree of relevance between the question received in the question receiving step and the predetermined consultation answer data identified in the identifying step is equal to or less than a predetermined value. This makes it possible to avoid generating an inappropriate answer based on the generated prompt when the content of the consultation answer data searched based on the question text has low relevance to the question text.

[0086] (Appendix 4) The program described in Appendix 2, wherein the identifying step (S103) includes a step of excluding one or more answers from the predetermined consultation answer data based on information indicating that one or more answers contained in the plurality of consultation answer data are appropriate answers to the question. This allows the AI ​​to obtain higher quality answers to questions received from users by excluding low-quality answers from the actual consultation response data that has been given in the past in response to prompts entered into the AI.

[0087] (Appendix 5) The program described in Appendix 2, wherein the identification step (S103) includes a step of excluding one or more answers from the predetermined consultation answer data based on evaluation information stored in association with the answer, from one or more answers included in the plurality of consultation answer data. This allows the AI ​​to obtain higher quality answers to questions received from users by excluding low-quality answers from the actual consultation response data that has been given in the past in response to prompts entered into the AI.

[0088] (Appendix 6) 6. The program of claim 5, wherein the evaluation information is information based on reactions given by a specified expert. This allows the AI ​​to obtain higher quality answers to questions received from users by excluding low-quality answers based on highly accurate evaluation information provided by experts from the actual consultation and response data from the past in response to prompts entered into the AI.

[0089] (Appendix 8) The program according to appendix 2, wherein the generating step (S104) is a step of generating a prompt including predetermined consultation answer data consisting of the question searched in the identifying step and the answer to the question. This allows the prompts input to the dialogue response AI to include data on actual consultation responses given in the past, making it possible to obtain higher quality answers to questions received from users.

[0090] (Appendix 9) The program described in Appendix 1, wherein the question receiving step (S101) is a step of receiving multiple questions from a user, and the identification steps (S102, S103) are steps of identifying predetermined consultation answer data based on a question that combines the multiple questions received in the question receiving step. This allows for obtaining higher quality answers based on questions that combine multiple questions.

[0091] (Appendix 10) A program described in any one of Appendices 1 to 9, in which a processor executes an output step (S105) in which the prompt generated in the generation step is output to an LLM (large-scale language model), a reception step (S106) in which an answer content from the LLM to the prompt output in the output step is received, and a presentation step (S107) in which the answer content received in the reception step is presented. By using LLM, it is possible to obtain higher quality answers to questions received from users.

[0092] (Appendix 11) A program as described in Appendix 10, in which a processor executes a card creation step (S108) of creating a legal consultation card for consulting with an expert based on the question received in the question receiving step and the answer received in the receiving step. This makes it easy to create a legal consultation card that is required when consulting with a specialist.

[0093] (Appendix 12) The program described in Appendix 11, wherein the card creation step (S108) is a step of creating a card based on text obtained by outputting the question received in the question receiving step and the answer received in the receiving step to the LLM. This makes it easy to create a legal consultation card that is required when consulting with a specialist.

[0094] (Appendix 13) The program described in Appendix 11, wherein the processor executes an inquiry step (S108) in which the processor inquires of multiple experts as to whether they can handle the consultation based on the legal consultation card created in the card creation step. This allows the user to easily find an expert who can respond to the user's question.

[0095] (Appendix 14) The program according to claim 10, wherein the presentation step (S107) is a step of presenting the answer content in association with at least a part of the predetermined consultation answer data identified in the identification step. This allows users to refer to the consultation and response data that was used as the source for generating the response content, along with the response content from LLM, allowing users to obtain more comprehensive information regarding their questions.

[0096] (Appendix 15) The program according to claim 14, wherein the presentation step (S107) is a step of not presenting a link for referring to predetermined consultation answer data. This makes it possible to prevent the user from referencing the consultation answer data that is the source when generating the answer content, along with the answer content from the LLM.

[0097] (Appendix 16) The program according to claim 14, wherein the presenting step (S107) is a step of presenting a link for referencing the predetermined consultation answer data in association with at least a part of the predetermined consultation answer data. This allows users to view the consultation response page that was used as the source for generating the response content, along with the response content from LLM, by opening the link. This allows users to view more comprehensive information regarding their questions.

[0098] (Appendix 17) The program according to claim 14, wherein the presenting step (S107) includes a step of presenting at least a part of the predetermined consultation answer data in association with information about the expert who provided the answer included in the predetermined consultation answer data. This allows users to view the response from LLM along with information about the expert who provided the answer, which serves as the source for generating the response, by opening the link. Users can easily find experts who can respond to their questions.

[0099] (Appendix 18) An information processing device comprising a processor and a memory unit, wherein the processor executes a question receiving step (101) of receiving a question from a user, a specification step (S102, S103) of identifying one or more predetermined consultation answer data from a plurality of consultation answer data consisting of questions or answers to past legal consultations based on the question received in the question receiving step, and a generation step (S104) of generating a prompt, which is an input sentence for a dialogue response AI, based on the question received in the question receiving step and the predetermined consultation answer data specified in the specification step. This allows the prompts input to the dialogue response AI to include data on actual consultation responses given in the past, making it possible to obtain higher quality answers to questions received from users.

[0100] (Appendix 19) A method for manufacturing an information processing system including an information processing device having a processor and a memory unit, the method comprising providing the information processing device with a program described in any one of Appendices 1 to 17. This allows the prompts input to the dialogue response AI to include data on actual consultation responses given in the past, making it possible to obtain higher quality answers to questions received from users.

[0101] (Appendix 20) An information processing method executed by a computer having a processor and a memory unit, wherein the computer executes a question receiving step (101) of receiving a question from a user, a specification step (S102, S103) of identifying one or more predetermined consultation answer data from a plurality of consultation answer data consisting of questions or answers to past legal consultations based on the question received in the question receiving step, and a generation step (S104) of generating a prompt, which is an input sentence for a dialogue response AI, based on the question received in the question receiving step and the predetermined consultation answer data specified in the specification step. This allows the prompts input to the dialogue response AI to include data on actual consultation responses given in the past, making it possible to obtain higher quality answers to questions received from users. [Explanation of symbols]

[0102] 1 System, 10 Server, 101 Memory Unit, 104 Control Unit, 106 Input Device, 108 Output Device, 20 User Terminal, 201 Memory Unit, 204 Control Unit, 206 Input Device, 208 Output Device, 50 LLM System, 501 Memory Unit, 504 Control Unit, 506 Input Device, 508 Output Device

Claims

1. A program to be executed by a computer having a processor and a storage unit, the processor: a question receiving step of receiving a question from a user; an identifying step of identifying one or more predetermined text data similar to the question from a plurality of text data related to law, based on the question received in the question receiving step; an input step of inputting input information including the question received in the question receiving step and the predetermined text data identified in the identifying step into a large-scale language model; an acquisition step of acquiring an answer content output from the large-scale language model in response to the input information; a presentation step of presenting the answer content acquired in the acquisition step; the presenting step does not present the answer content when a similarity between the question accepted in the question accepting step and the predetermined text data identified in the identifying step is equal to or less than a predetermined value. program.

2. the identifying step is a step of identifying the predetermined text data by searching the plurality of text data based on the question accepted in the question accepting step. The program according to claim 1.

3. the identifying step includes a step of excluding one or more text data from the predetermined text data based on evaluations associated with the plurality of text data and stored. The program according to claim 2.

4. the identifying step includes a step of excluding one or more pieces of text data from the predetermined text data based on evaluation information stored in association with the plurality of pieces of text data; The program according to claim 2.

5. The evaluation information is information based on reactions given by predetermined legal experts. The program according to claim 4.

6. A program to be executed by a computer having a processor and a storage unit, the processor: a question receiving step of receiving a question from a user; an identifying step of identifying one or more predetermined text data similar to the question from a plurality of text data related to law, based on the question received in the question receiving step; an input step of inputting input information including the question received in the question receiving step and the predetermined text data identified in the identifying step into a large-scale language model; an acquisition step of acquiring an answer content output from the large-scale language model in response to the input information; a presentation step of presenting the answer content acquired in the acquisition step; an information creation step of creating information for consulting with a legal expert based on the question received in the question receiving step and the answer content acquired in the acquisition step; To execute program.

7. the information creation step is a step of creating the information based on information obtained by inputting the question accepted in the question acceptance step and the answer content acquired in the acquisition step into a large-scale language model. The program according to claim 6.

8. the processor: an inquiry step of inquiring of a plurality of legal experts as to whether they can handle the consultation based on the information created in the information creation step; To execute The program according to claim 6.

9. A program to be executed by a computer having a processor and a storage unit, the processor: a question receiving step of receiving a question from a user; an identifying step of identifying one or more predetermined text data similar to the question from a plurality of text data related to law, based on the question received in the question receiving step; an input step of inputting input information, including the question received in the question receiving step and the predetermined text data identified in the identifying step, into a large-scale language model; an acquisition step of acquiring an answer content output from the large-scale language model in response to the input information; a presentation step of presenting the answer content acquired in the acquisition step; Run the presenting step is a step of presenting the answer content in association with at least a part of the predetermined text data identified in the identifying step. program.

10. the presenting step is a step of not presenting a link for referencing the predetermined text data. The program according to claim 9.

11. the presenting step is a step of presenting a link for referencing the predetermined text data in association with at least a part of the predetermined text data. The program according to claim 9.

12. the presenting step includes a step of presenting at least a portion of the predetermined text data in association with information about the legal expert who provided the answer, the information being included in the predetermined text data. The program according to claim 9.

13. A method executed by an information processing device including a processor and a storage unit, wherein the processor executes all of the steps executed in the invention according to any one of claims 1 to 12.

14. 13. An information processing device comprising a processor and a storage unit, wherein the processor executes all of the steps executed in the invention according to any one of claims 1 to 12.

15. A system comprising means for executing all steps performed in an invention according to any one of claims 1 to 12.

Images