DEVELOPMENT DOCUMENT CREATION DEVICE, DEVELOPMENT DOCUMENT CREATION METHOD, AND DEVELOPMENT DOCUMENT CREATION PROGRAM

Abstract

A development document creation device, a development document creation method, and a development document creation program are provided that enable the creation of documents specialized for system development. [Solution] The development document creation device of the present invention comprises a development information acquisition unit that acquires development information, a first recording unit that records a prompt of a large-scale language model as template information according to a development task, a task acquisition unit that acquires the development task from a user, a question generation unit that generates a first question based on the template information according to the acquired task, an answer acquisition unit that acquires a first user answer to the first question from the user, an answer generation unit that generates an answer that is a solution based on the first user answer, and a creation unit that creates a development document based on the development information and the first user answer to the first question acquired from the user.

Description

[Technical field]

[0001] The present invention relates to a development document creation device, a development document creation method, and a development document creation program, and in particular to a development document creation device, a development document creation method, and a development document creation program that are capable of supporting the creation of development documents using a large-scale language model. [Background technology]

[0002] Conventionally, there is a technology that enables easy ordering activities by providing sufficient support throughout the entire ordering process. For example, an ordering support device disclosed in Patent Document 1 has a large-scale language model section having a large-scale language model specialized for creating ordering support information by probabilistically predicting how likely words and sentences given in a prompt are to occur as natural language based on ordering-related information in an ordering database, and learning and adjustment are performed to generate ordering support information, an information receiving section that receives user response information in conversation with the user, an initial setting prompt generating section that instructs as a precondition for generating ordering support information in the large-scale language model, and an ordering prompt generating section that causes the large-scale language model to generate ordering support information suitable for the ordering stage based on the precondition.

[0003] However, the order placement and receipt support device disclosed in Patent Document 1 utilizes a large-scale language model specialized for creating order placement and receipt support information, and focuses on holding conversations with users according to the order placement and receipt stages from receiving the user's task to placing an order with the vendor, but does not focus on creating documents specialized for system development, such as in-house proposals, environment definition documents, requirements definition documents, design documents, and plans. For this reason, there was a demand for technology that would enable the creation of development-specific documents, such as in-house proposals, environment definition documents, requirements definition documents, design documents, and plans, in system development. [Prior art documents] [Patent documents]

[0004] [Patent Document 1] Patent No. 7526415 Summary of the Invention [Problem to be solved by the invention]

[0005] Therefore, the present invention has been made in consideration of the above points, and provides a development document creation device, a development document creation method, and a development document creation program that enable the creation of development-specific documents such as in-house proposals, environment definition documents, requirements definition documents, design documents, and plans in system development. [Means for solving the problem]

[0006] That is, the development document creation device of the first aspect is a development document creation device that supports the creation of development documents using a large-scale language model, and is characterized by comprising: a development information acquisition unit that acquires development information related to development to be described in the development document, a first recording unit that records a prompt of the large-scale language model as template information according to the development task, a task acquisition unit that acquires the development task from a user, a question generation unit that generates a first question using the large-scale language model based on the template information according to the acquired task, an answer acquisition unit that acquires a first user answer to the first question from the user, an answer generation unit that generates an answer that is a solution based on the first user answer acquired by the answer acquisition unit, and a creation unit that creates development documents using the large-scale language model based on the development information and the first user answer to the first question acquired from the user.

[0007] In a second aspect, in the development document creation device of the first aspect, the development documents include an in-house proposal, an environment definition document, a requirements definition document, a design document and a plan, and the creation unit creates an in-house proposal, creates a current business flow, a list of issues and an environment definition document based on the in-house proposal, creates a requirements definition document based on the current business flow, the list of issues and the environment definition document, and creates a design document and a plan based on the created requirements definition document.

[0008] In the third aspect of the development document creation device, the first recording unit may record enterprise resource planning implementation procedures and maintenance know-how, and the creation unit may create development documents corresponding to the recorded enterprise resource planning implementation procedures and maintenance know-how as an output.

[0009] In the development document creation device of the fourth aspect, the creation unit may create the development document based on a first prompt that is a pre-templated enterprise resource planning implementation work prompt, and a second prompt based on a dialogue with the user.

[0010] The development document creation device of the fifth aspect may further include a second recording unit that stores a prompt created by the creation unit and used to create the development document, a shared display unit configured to enable other users in the same team to view and execute a project in which one user is in charge, and a reflection unit that reflects the prompt when creating another development document different from the development document.

[0011] In the development document creation device of the sixth aspect, the question generator may further generate a second question that complements the first question based on a first user answer from the user.

[0012] In a development document creation device according to a seventh aspect, the answer acquisition unit acquires a second user answer to the second question from the user; The creation unit may create the development document based on the development information, the first response, and the second response.

[0013] In the development document creation device of the eighth aspect, the first prompt may include a conversation history between the user and the AI ​​required for creating the development document.

[0014] In the development document creation device of the ninth aspect, the answer generating unit may generate an answer corresponding to a category of the task.

[0015] In the development document creation device of the tenth aspect, the answer generation unit may generate an answer according to a current date.

[0016] A development document creation method of an eleventh aspect is a method for supporting the creation of development documents using a large-scale language model, characterized in that it has a computer execute the following steps: a development information acquisition step for acquiring development information related to development to be described in the development document; a recording step for recording a prompt of the large-scale language model as template information according to the development task; a task acquisition step for acquiring a development task from a user; a question generation step for generating a first question using the large-scale language model based on the template information according to the acquired task; an answer acquisition step for acquiring a first user answer to the first question from the user; an answer generation step for generating an answer that is a solution based on the first user answer acquired by the answer acquisition unit; and a creation step for creating development documents using the large-scale language model based on the development information and the first user answer to the first question acquired from the user.

[0017] A development document creation program according to a twelfth aspect is a program for supporting the creation of development documents using a large-scale language model, and is characterized in that the development document creation program has a computer realizing a development information acquisition function for acquiring development information related to development to be described in the development document, a recording function for recording a prompt of the large-scale language model as template information according to the development task, a task acquisition function for acquiring the development task from a user, a question generation function for generating a first question using the large-scale language model based on the template information according to the acquired task, an answer acquisition function for acquiring a first user answer from the user to the first question, an answer generation function for generating an answer that is a solution based on the first user answer acquired by the answer acquisition unit, and a creation function for creating development documents using the large-scale language model based on the development information and the first user answer to the first question acquired from the user. Effect of the Invention

[0018] The development document creation device of the present invention is a development document creation device that supports the creation of development documents using a large-scale language model, and is characterized in that it comprises a development information acquisition unit that acquires development information related to development to be described in the development document, a first recording unit that records a prompt of the large-scale language model as template information according to the development task, a task acquisition unit that acquires the development task from a user, a question generation unit that generates a first question using the large-scale language model based on the template information according to the acquired task, an answer acquisition unit that acquires a first user answer to the first question from the user, an answer generation unit that generates an answer that is a solution based on the first user answer acquired by the answer acquisition unit, and a creation unit that creates development documents using the large-scale language model based on development information and the first user answer to the first question acquired from the user, thereby enabling the creation of documents specialized for development, such as in-house proposals, environment definition documents, requirements definition documents, design documents and plans, in system development.

[0019] In addition, the development document creation method and development document creation program of the present invention, like the development document creation device of the present invention, enable the creation of development-specific documents in system development, such as in-house proposals, environment definition documents, requirements definition documents, design documents and plans. [Brief description of the drawings]

[0020] [Figure 1] FIG. 1 is a diagram showing an outline of the configuration of a development documentation system according to an embodiment of the present invention. [Diagram 2] FIG. 2 is a block diagram showing the functional configuration of the development document creation device according to the embodiment of the present invention. [Diagram 3] FIG. 3 is a diagram showing the specifications of the development document creation device according to the embodiment of the present invention. [Figure 4] FIG. 4 is a diagram showing screen transitions in the development document creation device according to the embodiment of the present invention. [Diagram 5] FIG. 5 is a diagram showing a chat screen of the development document creation device according to the embodiment of the present invention. [Figure 6]FIG. 6 is a diagram showing a screen image of the development document creation device according to the embodiment of the present invention. [Figure 7] FIG. 7 is a diagram showing a template of the development document creation device according to the embodiment of the present invention. [Figure 8] FIG. 8 is a diagram showing a project template of the development document creation device according to the embodiment of the present invention. [Figure 9] FIG. 9 is a diagram showing a process template of the development document creation device according to the embodiment of the present invention. [Figure 10] FIG. 10 is an example of a flowchart of a development document creation method according to an embodiment of the present invention. DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS

[0021] Hereinafter, an embodiment of the present invention will be described with reference to the drawings. Note that the drawings are merely examples, and the present invention is not limited to those shown in the drawings. For example, the development document creation device (server), large-scale language model server, and number of user terminals, data set (table), flow chart, and display screen example shown in the drawings are merely examples, and the present invention is not limited to these.

[0022] <System configuration> 1 is a diagram showing an example of the configuration of a development document creation system according to an embodiment of the present invention. The development document creation system 600 includes at least a server (development document creation device) 100, a user terminal 200, and a large-scale language model server 300.

[0023] The server 100 functions as a platform that provides a development document creation support service, and outputs the created development documents to the user terminal 200 connected via the network 500. Here, the "development documents" are documents specialized for development in system development, and may refer to in-house proposals, environment definition documents, requirements definition documents, design documents, plans, etc. As will be described in detail later, the server 100 creates the development documents while exchanging information with the user terminal 200. At this time, the server 100 creates the development documents using the large-scale language model functions provided by the large-scale language model server 300 in the cloud.

[0024] Large language models (LLM) are a type of generative AI (Artificial Intelligence) specialized in natural language processing (NLP), and are language models that realize advanced natural language generation (NLG) constructed by deep learning of a huge amount of text data. Note that a language model refers to a mechanism that understands the patterns of natural language used by humans and predicts the next word that comes after a certain word by analyzing the occurrence probability of the word. Examples of large language models may be ChatGPT (Chat Generative Pre-trained Transformer) (registered trademark), PaLM2 (registered trademark), LaMDA2 (registered trademark), etc. Note that words and sentences used as instructions to be input to these large language models are generally called prompts. Note that the language model used is not limited to a large language model, and may be a small language model (SLM). A small language model is a lightweight language model that excels in processing specific tasks. When users input files (attached files), the system supports text formats (txt, html, xml, csv), Office files (Excel, PowerPoint, Word), PDF, and images (jpg, png). These files are converted to text format and saved as conversation history linked to the user's remarks. For images in particular, the system is set up to obtain two types of information: "OCR result information" and "information converted from the image into text format," and store these in the conversation history, allowing the AI ​​to provide more accurate answers even for images that contain text.

[0025] The server 100 may be any device as long as it is an information processing device capable of realizing the functions described in each embodiment. The server 100 includes, for example, a server device, a computer (for example, a desktop, a laptop, a tablet, etc.), or a communication platform. In FIG. 1, the server 100 and the large-scale language model server 300 are shown as separate entities, but this is not limited thereto. That is, each function described as being provided by the server 100 may be realized by a plurality of servers, or may be realized by a single server. In addition, the function provided by the large-scale language model server 300 may be realized as a part of the function of the server 100. The server 100 may be, for example, a distributed server system that cooperates by communicating via a network, or may be a so-called cloud server. That is, the server 100 is not limited to a physical server, and may also include a virtual server by software. In addition, the server 100 may be any device as long as it is an information processing device capable of realizing the functions described in each embodiment, and may include, for example, a server device, a computer (for example, but not limited to, a desktop, a laptop, a tablet, etc.), a communication platform, etc.

[0026] The user terminal 200 is a communication terminal used by a user of the development documentation creation support system 600, and has an application for using the development documentation creation support service installed thereon, and is capable of transmitting and receiving various types of information to and from the server 100. Note that it is not essential to install an application on the user terminal 200, and the user may access a web page for using the development documentation creation support service provided by the server 100 from the user terminal 200 via a web browser or the like, and transmit and receive various types of information to and from the server 100.

[0027] 1 shows a notebook personal computer as the user terminal 200, any terminal capable of implementing the functions described below may be used as the user terminal 200. For example, the user terminal 200 may be a computer (e.g., a tablet terminal, a desktop personal computer), a smartphone, etc.

[0028] The development documentation creation support system 600 may further include a database server (not shown) that stores various information (data) used in the development documentation creation support system 600. The data used in the development documentation creation support system 600 may be, for example, a prompt to be input to a large-scale language model or user information.

[0029] The network 500 may include a wireless network or a wired network, and may be, for example, a wireless LAN (WLAN), a wide area network (WAN), integrated service digital networks (ISDNs), wireless LANs, code division multiple access (CDMA), long term evolution (LTE), LTE-Advanced, fourth generation communication (4G), fifth generation communication (5G), sixth generation communication (6G) or later mobile communication systems. Note that the network 500 is not limited to these examples, and may be, for example, a public switched telephone network (PSTN), Bluetooth (registered trademark), optical lines, a satellite communication network, or the like. The network 500 may also be a combination of these.

[0030] 2 is a functional block diagram of a development document creation device (server) 100 according to an embodiment. The development document creation device 100 according to this embodiment includes a development information acquisition unit 10, a first recording unit 20, a task acquisition unit 30, a question generation unit 40, an answer acquisition unit 50, an answer generation unit 60, a creation unit 70, a second recording unit 80, a shared display unit 90, and a reflection unit 110. These are functional units that are realized by a CPU executing an information processing program deployed on a RAM. The development document creation device 100 further includes storage units such as a ROM, a RAM, an SSD, and an HDD, various interfaces, and the like, appropriately implemented (not shown).

[0031] The development information acquisition unit 10 acquires development information related to development to be described in development documents. Development documents include, for example, in-house planning documents, environment definition documents, requirements definition documents, design documents, and plans. The development information is information such as enterprise resource planning (ERP) implementation know-how, ERP maintenance know-how, standard documents, ASIS business materials for customers, security masks, etc. Security masks are a technology that uses data registered in a mask keyword master to protect the confidentiality of data by hiding, rewriting, or changing specific information while maintaining the original data structure and meaning.

[0032] The first recording unit 20 records the prompt of the large-scale language model as template information according to the development task. The development task is, for example, creating a requirements definition document, creating a list of new functions, creating a basic design document, creating a detailed design document, programming, etc. By including the development information acquisition unit 10 and the first recording unit 20, the development document creation device 100 can create development documents by statistically processing at the word level in response to specialized know-how. This enables work that is not dependent on the user's ERP specialized knowledge or AI skills.

[0033] The task acquisition unit 30 acquires development tasks from the user. As described above, development tasks include, for example, creating a requirements definition document, creating a list of new functions, creating a basic design document, creating a detailed design document, programming, and the like, and the user selects the work he or she wants to perform from a template. The task acquisition unit 30 acquires tasks related to the template selected by the user. Since the work prompts for ERP implementation are templated in advance, once a task is selected from the template, questions are asked by the AI ​​and the user can start working.

[0034] The question generation unit 40 generates a first question using a large-scale language model based on the acquired template information corresponding to the task. The question generation unit 40 further generates a second question that complements the first question based on the first user answer from the user. The first question is, for example, "Please tell me the problem your company is facing," "What problems does the current system and business have," "If you know the cause, please tell us as well," and the like. The first user answer is, for example, "It is taking a long time to collect revenue for revenue management by department and segment. I think that is because there is no system," "We are unable to grasp the cost of individual products. This is because we are unable to allocate common costs," "We are unable to grasp the yield," "Approval processing is paper-based," "Legacy systems remain," "The maintenance deadline for the ERP we are using is approaching," and the like. The second question is, for example, "You said that it takes a long time to collect revenues, but what is the specific process for collecting revenues? Also, which department or segment is particularly problematic?", "You said that you are unable to allocate common expenses, but how do you currently calculate costs? What kind of standards do you want to set for the method of allocating common expenses?", "You said that you are unable to grasp the yield, but how do you currently manage production data? How do you collect the data required for calculating the yield?", "You said that the approval process is paper-based, but what approval process exists? Also, how are the departments and personnel involved in the approval involved?", "You said that a legacy system remains, but what system specifically remains and what kind of work is it used for? Also, which part of the legacy system is causing the problem?", etc. From then on, the question generation unit 40 digs deeper and deeper to generate questions such as the third question, the fourth question, and so on.

[0035] The answer acquisition unit 50 acquires a first user answer to the first question from the user. The answer acquisition unit 50 acquires a second user answer to the second question from the user. Examples of the second user answer include: "It's email exchanges and meetings. It takes about 20 hours / month in total. The problem is with the A department.", "The cost is calculated with the XX system, but I don't know the details (I) unless I check with the person in charge of the package. I need to check.", "Production management uses the YY system. Data collection is done manually.", "There are many approval processes from subordinates to superiors. However, there are also flows that require final approval by the information systems department or management department.", "We still have an inventory management system. It is an independent system and does not have sufficient integration with other current systems, which is a problem.", "The support deadline is in September 2025, and we are in trouble because the vendor will not provide support after the deadline.", etc.

[0036] The answer generating unit 60 generates an answer that is a solution based on the first user answer acquired by the answer acquiring unit. Even if the user gives an irrelevant answer, an answer that is a solution is generated.

[0037] The answer generating unit 60 generates answers such as a third answer, a fourth answer, and so on. The third answer and onwards may be the "ideal state." Examples of the "ideal state" include "utilizing the revenue management information of the software XX to build an automated revenue collection process," "using the cost management function of the software XX to set the allocation standard for common expenses," "utilizing the production management function of the software XX to automatically collect data required for yield calculation," "using the workflow function of the software XX to build an electronic approval process," "planning a full-scale transition to the software XX and starting integration," "promptly proceeding with the transition to the software XX and avoiding maintenance risks," and so on.

[0038] The answer generation unit 60 generates answers according to the task category. In order to stabilize the behavior of the AI, a mechanism is installed that recognizes what genre the current task is. Specifically, the "categories" installed in general systems are expanded and linked to system prompts to the AI. This allows the AI ​​to correctly recognize the topic and behave appropriately. When a category is set, the AI ​​is automatically instructed to behave as a specialist in that category. It is also possible to set multiple categories. The system prompts automatically given based on the category cause the AI ​​to behave as a specialist in software XX. Without this specification, the AI ​​will start talking about general systems.

[0039] The answer generation unit 60 generates an answer according to the current date. Since AI usually has the tendency to recognize the knowledge cutoff date as the current date, the current date is correctly recognized by providing current date information within the program. Even if the user makes a comment without providing date information, the AI ​​correctly understands the date, so the entry date is correctly entered.

[0040] The answer generation unit 60 uses the concept of RAG and is equipped with a mechanism for answering by taking into account search results from search resources. RAG stands for Retrieval-Augmented Generation, and is a framework that improves output accuracy that is difficult to control with prompts alone by combining text generation from a large-scale language model (LLM) with a search for reliable external information. By having the AI ​​select the search target, problems such as a decrease in answer accuracy due to unnecessary information and an excess of tokens (speed reduction, errors) due to unnecessary information are solved. However, this alone can also cause a problem in which the AI ​​does not execute the search that you want it to execute. For this reason, the development document creation device 100 makes it possible to clearly specify the searches that you want it to execute and the parts that you want it to judge.

[0041] As a way to obtain more important data in the graph DB, a large number of routes are generally found when nothing is specified when searching for routes in the graph DB. On the other hand, since the number of tokens is limited in the AI, it is necessary to obtain only important routes. The development document creation device 100 is equipped with a mechanism that gives attribute information related to importance to edge property information and examines the searched information (i.e., obtains only information with high importance). When a question that requires a search in the graph DB is asked, the AI ​​judges that "a search in the graph DB is necessary," executes a search process in the graph DB internally, and outputs an answer that takes into account the search results of the graph DB. It can also clearly answer intermediate table relations. This is content that cannot be answered by simple text search or vector search. For example, when searching for a route between A and C for something related to A → B → C, such as a table relation, there was a problem that it was difficult to obtain data corresponding to B with conventional text search or vector search. In the development document creation device 100, the intended answer can be obtained by storing data in the graph DB and performing RAG from there to solve this problem. This makes it possible, for example, to easily search for relationships between tables included (pre-bundled) in package software such as ERP, allowing AI to write the correct queries.

[0042] The creation unit 70 creates development documents using a large-scale language model based on the development information and the first user answer to the first question acquired from the user. The creation unit 70 creates an in-house planning document, creates a current business flow, a list of issues, and an environment definition document based on the in-house planning document, creates a requirements definition document based on the current business flow, the list of issues, and the environment definition document, and creates a design document and a plan document based on the created requirements definition document. The creation unit 70 creates development documents corresponding to the recorded enterprise resource planning (ERP) implementation procedure and ERP maintenance know-how as deliverables. The creation unit 70 creates development documents based on a first prompt (back prompt) that is a template of an ERP implementation work prompt in advance, and a second prompt (front prompt) based on a dialogue with the user. The creation unit 70 creates development documents based on the development information, the first answer, and the second answer. The first prompt (back prompt) includes a conversation history between the user and the AI ​​that is necessary for creating development documents. The conversation is a dialogue between the user and the AI, and there may be multiple users. In other words, for example, examples could include user A's comments, AI's comments, user B's comments, and AI's comments.

[0043] The second recording unit 80 stores the prompts used to create the development documents created by the creation unit 70. The prompts used to create the development documents created by the creation unit 70 are prompts from the previous process, such as the ASIS function list, the new system requirements definition document, the new system non-functional requirements, the issue list, the organization / person in charge list, and the authority configuration list. These previous process prompts and the development documents as deliverables can be created by the AI ​​responding to them and statistically processing them at the word level. For example, if the current process is a task to create a basic design document, the requirements definition prompt, the latest requirements definition document, the issue list, etc. are referenced, and a basic design task for the add-on function is started to create the basic design document.

[0044] The shared display unit 90 is configured so that other users in the same team can view and execute a project that one user is in charge of. On the home screen, it is possible to display not only the project that one is working on, but also the projects of other members. This makes it possible, for example, for a veteran employee to take over the work of a new employee and start working from wherever it left off, or for a manager to check the status of each member's work. If the filter narrowed down by the logged-in user is released, projects of users other than the user will be displayed and can be viewed and executed. Since it is equipped with a team member sharing mode, work can be shared between members. In addition, AI can refer to and utilize all information in the project.

[0045] The reflecting unit 110 reflects the prompt when creating a development document other than the development document. For example, if the project in the previous process is the creation of a requirements definition document, the AI ​​can respond to prompts such as the ASIS function list, new system requirements definition document, new system non-functional requirements, list of issues, list of organizations and personnel, and list of authority configurations, and the requirements definition document, which is the development document that is the deliverable, and statistically process them at the word level to create them. Then, for example, if the other development document to be created in the current process is a basic design document, the AI ​​can refer to the requirements definition prompt, the latest requirements definition document, the list of issues, etc., start the basic design task for the add-on function, and create the basic design document.

[0046] FIG. 3 is a diagram showing the specifications of the development document creation device 100 according to the embodiment of the present invention. As the introduction process template 1, a requirements definition template 11, a function list template 12, an additional development basic design template 13, and an additional development program template 14 are prepared. The worker 2 selects a template suitable for the work to be performed from these templates. Meanwhile, the development document creation device 100 acquires enterprise resource planning (ERP) introduction knowledge and system prompts 3. The enterprise resource planning (ERP) introduction knowledge and system prompts 3 include ERP introduction know-how 31, ERP standards 32, and document templates 33. The development document creation device 100 acquires these ERP introduction know-how 31, ERP standards 32, and document templates 33, as well as web latest information 34. Then, upon receiving a consultation or request from the worker 2, the development document creation device 100 prompts for necessary information or executes the work. Then, the development document creation device 100 navigates the worker 2 and provides the worker 2 with a deliverable 4. The deliverable 4 includes a requirements definition document 41, a design document 42, and a source code 43. Source code is a string of characters (text or text files) written in a programming language that represents a computer program.

[0047] FIG. 4 is a diagram showing screen transitions of the development document creation device 100 according to the embodiment of the present invention. The left side shows the execution environment 5, and the right side shows the server setting environment 6. When logging in to the execution environment 5, a project list screen 51 is displayed. When the worker 2 selects the execution screen 52 on the project list screen 51, the execution screen 52 is displayed, and when the worker 2 selects the setting screen 53, the setting screen 53 is displayed. On the setting screen 53, a mask keyword registration 54 can be performed. A mask keyword is a keyword that is hidden so that it can be registered so that other users cannot know it. For example, when personal information such as "Yamada Hanako" is mistakenly entered in the chat field, this is a master function for automatically converting it to another character string "XX XX" and then sending it to the AI. In this case, a pair of "Yamada Hanako" and "XX XX" is registered in advance in the mask keyword master.

[0048] When worker 2 logs in to server setting environment 6, management screen 61 is displayed. On management screen 61, worker 2 can make various registrations. The various registrations are project template registration 62, process registration 63, task registration 64, category registration 65, disclosure destination registration 66, and license registration 67. In project template registration 62, the template of each project can be registered. In process registration 63, each process can be registered. In task registration 64, each task can be registered. In category registration 65, each category can be registered. For disclosure destination registration 66 and license registration 67, first the disclosure destination (disclosure destination A, disclosure destination B, ...) and license (license X, license Y, ...) are registered, and then each license is assigned to the disclosure destination.

[0049] With reference to FIG. 5, a chat screen of the development document creation device according to the embodiment of the present invention will be described. FIG. 5 is a diagram showing a chat screen of the development document creation device according to the embodiment of the present invention. The chat screen shown in FIG. 5 shows a state of hearing for creating a problem list. The upper part shows questions from the AI, and the middle part shows answers from the user. For example, as shown in the upper part of FIG. 5, the questions from the AI ​​are summarized as questions such as "Please tell us the problems your company is facing," "What problems are there in the current system and business?", and "If you know the cause, please also tell us." The answers from the user to those questions are summarized as questions such as "We are not able to manage income and expenditure by construction project," "The current system developed from scratch can show profit and loss by organization, but does not have a function to manage by project, so the profit and loss of the project is unknown," and "In the new system, we would like to use standard functions to aggregate sales and costs by project and manage the profit and loss of each project." The AI's questions in the upper row and the user's answers in the middle row do not necessarily correspond one-to-one, and the user can also answer the user's wishes, such as "I want the new system to use standard functions to aggregate sales and costs by project and manage the profit and loss of each project." The AI ​​can respond to what the user wants to write, and can propose an answer that takes that into account, that is, the "ideal state." The lower row shows the first answer to the AI's question, "We are not able to manage the income and expenses of each construction project," and the second answer is about to be entered as a message in the comments field. The user can also upload a file when answering. If the user is finished with this second answer, they press the Complete button in the lower right to proceed to the next task, and if they want to enter a third answer, they press the Continue button in the lower right to enter their answer.

[0050] FIG. 6 is a diagram showing a screen image of the development document creation device 100 according to the embodiment of the present invention. FIG. 6 is an image of a project list screen 7. First, when "Display only projects created by me" is checked, only projects created by me are displayed. The first column from the left displays the ID 71, the second column from the left displays the project name 72, the third column from the left displays the template 73, the fourth column from the left displays the target category 74, the fifth column from the left displays the creator 75, and the sixth column from the left displays the creation date 76. These items can be paged and sorted. Clicking the "+" button pops up a new project creation screen 8. After changing the template 81 selection, pressing the Confirm button 82 will display the related processes. The first column from the left displays the sequence 83, the second column from the left displays the process ID 84, and the third column from the left displays the project name 85.

[0051] 7 is a diagram showing templates of the development document creation device 100 according to the embodiment of the present invention. Process templates 91-1 to 91-3 are linked under a project template 91, and each of the process templates 91-1 to 91-3 includes a template for each task. For example, as shown in FIG. 7, the process template 91-1 includes tasks 1-1 to 1-3 as templates for task 1. The process template 91-2 includes tasks 2-1 and 2-2 as templates for task 2.

[0052] FIG. 8 is a diagram showing a project template of the development document creation device 100 according to the embodiment of the present invention. In FIG. 8, a project template 91 is shown. In this case, a screen for creating a problem list with ID 13 is displayed. A license and a category are displayed. In the table in the center, the vertical axis is a number, and the horizontal axis is, from the left, a process ID, a process name, and a process category. For example, as shown in FIG. 8, No. 1 is the creation of a problem list with process ID 31, and the process category is software xx. No. 2 is the consideration of a problem solution with process ID 32, and the process category is software xx. No. 3 is the output of a problem list file with process ID 37, and no process category is assigned. These orders can be changed using an up arrow and a down arrow. For example, when the process name "creation of a problem list" is clicked, the process template shown in FIG. 9 is displayed. Note that the screen shown in FIG. 8 is not a screen used by a user, but a screen used by an administrator for template setting.

[0053] FIG. 9 is a diagram showing a process template of the development document creation device 100 according to the embodiment of the present invention. In FIG. 9, a process template 91 is shown. In this case, "creation of a task list" of process ID 31 is displayed. The category is software XX. In the table in the center, the vertical axis is a number, and the horizontal axis is, from the left, a task ID, a task name, and a task category. For example, as shown in FIG. 8, No. 1 is confirmation of a task of task ID 86, and the task category is software XX. No. 2 is creation of a task list of task ID 87, and the task category is software XX. No. 3 is summary of a task list of task ID 88, and the task category is software XX. These orders can be changed using the up arrow and the down arrow. Note that the screen shown in FIG. 9 is not a screen used by a user, but a screen used by an administrator for template setting.

[0054] The development documentation creation program according to the embodiment will be described together with the development documentation creation method with reference to Fig. 10. Fig. 10 is an example of a flowchart of the development documentation creation method according to the embodiment. The development document creation method is executed by the CPU of the development document creation device 100 based on a development document creation program. The development document creation program includes a development information acquisition step S10, a recording step S20, a task acquisition step S30, a question generation step S40, an answer acquisition step S50, an answer generation step S60, and a creation step S70. The development document creation program causes the CPU of the development document creation device 100 to realize a development information acquisition function, a recording function, a task acquisition function, a question generation function, an answer acquisition function, an answer generation function, and a creation function. These functions are executed in the order shown in the flowchart of Fig. 9, but the order can be changed as appropriate. Note that each function overlaps with the description of the various functional units of the development document creation device 100 described above, and therefore detailed description thereof will be omitted.

[0055] The development information acquisition function acquires development information relating to the development to be described in the development document (Step S10: development information acquisition step).

[0056] The recording function records prompts of a large-scale language model as template information according to a development task (step S20: recording step).

[0057] The task acquisition function acquires a development task from a user (step S30: task acquisition step).

[0058] The question generation function generates a first question using a large-scale language model based on template information corresponding to the acquired task (step S40: question generation step).

[0059] The answer acquisition function acquires a first user answer to a first question from a user (Step S50: answer acquisition step).

[0060] The answer generating function generates an answer, which is a solution, based on the first user's answer acquired by the answer acquiring function (step S60: answer generating step). Note that the process from question generating step S40 to answer generating step S60 is not carried out just once, but is usually repeated 1 to N times. Once an answer is generated in answer generating step S60, the process returns to question generating step S40, a second question is generated, the second user's answer is acquired in answer acquiring step S50, and an answer is generated in answer generating step S60. From the second time onwards, the AI's answer and the next question may be output at the same time. The user decides whether to proceed from answer generating step S60 to the next creation step S70 or to return to question generating step S40.

[0061] The creation function creates a development document using a large-scale language model based on the development information and the first user answer to the first question acquired from the user (step S70: creation step).

[0062] According to each aspect of the present disclosure described above, it becomes possible to create documents specialized for development, such as in-house proposals, environment definition documents, requirements definition documents, design documents, and plans, in system development.

[0063] The development document creation program of the embodiment can be implemented in a CPU using a computer language such as Arduino (registered trademark) IDE.

[0064] [Functions and circuits] Next, the functions and circuits of the development document creation device 100 will be described. Each functional unit of the CPU of the development document creation device 100 may be realized as a function of a computer's arithmetic processing unit, etc. That is, the CPU may be realized as a development information acquisition function, a recording function, a task acquisition function, a question generation function, an answer acquisition function, an answer generation function, and a creation function by the computer's arithmetic processing unit, etc. The development document creation program can cause a computer to realize each of the above-mentioned functions. The development document creation program may be recorded in a non-transitory computer-readable storage medium, such as a memory, a solid state drive, a hard disk drive, or an optical disk. The storage medium may be rephrased as a non-transitory computer-readable medium that stores the development document creation program. The development document creation program may also be transmitted online. Furthermore, the above-mentioned computer arithmetic processing unit etc. may be configured, for example, by an integrated circuit etc. That is, the CPU of the development document creation device 100 may be realized as a development information acquisition circuit, a recording circuit, a task acquisition circuit, a question generation circuit, an answer acquisition circuit, an answer generation circuit, and a creation circuit that configure the computer arithmetic processing unit etc.

[0065] The present invention is not limited to the development document creation device 100, the development document creation method, and the development document creation program according to the above-described embodiments, and can be embodied in various other modified or applied examples without departing from the gist of the present invention described in the claims. Also, in the above-described embodiment, the word "information" is used, but the word "information" can be replaced with "data", and the word "data" can be replaced with "information".

[0066] Although the development document creation device 100 is used in this embodiment, it is not limited to development documents, and the creation, programming, and review of documents other than development documents are also possible, and there is no particular limit to the uses.

[0067] [Aspects and Effects of the Present Embodiment] Next, one aspect of this embodiment and the effects of each aspect will be described. Note that each aspect described below is an example at the time of filing, and this embodiment is not limited to the aspects described below. In other words, this embodiment is not limited to each aspect described below, and may be realized by appropriately combining each of the above-mentioned parts. In addition, a lower aspect may be able to cite any of the higher aspects. In addition, the effects of the present embodiment described below are merely examples, and the effects of each aspect are not limited to those described below. In addition, each aspect may, for example, achieve at least one of the effects described below.

[0068] (Aspect 1) One embodiment of a development document creation device is a development document creation device that supports the creation of development documents using a large-scale language model, and includes a development information acquisition unit that acquires development information related to development to be described in the development document, a first recording unit that records a prompt of the large-scale language model as template information according to the development task, a task acquisition unit that acquires the development task from a user, a question generation unit that generates a first question using the large-scale language model based on the template information according to the acquired task, an answer acquisition unit that acquires a first user answer to the first question from the user, an answer generation unit that generates an answer that is a solution based on the first user answer acquired by the answer acquisition unit, and a creation unit that creates development documents using the large-scale language model based on the development information and the first user answer to the first question acquired from the user. This enables the development document creation device to create documents specialized for development, such as in-house proposals, environment definition documents, requirements definition documents, design documents, and plans for system development. The same applies to the development documentation method and the development documentation program.

[0069] (Aspect 2) In one embodiment of the development document creation device, the development documents include an in-house proposal, an environment definition document, a requirements definition document, a design document and a plan, and the creation unit creates an in-house proposal, creates a current business flow, a list of issues and an environment definition document based on the in-house proposal, creates a requirements definition document based on the current business flow, the list of issues and the environment definition document, and creates the design document and the plan based on the created requirements definition document. This enables the development document creation device to create documents specialized for development, such as in-house proposals, environment definition documents, requirements definition documents, design documents, and plans for system development.

[0070] (Aspect 3) In one embodiment of the development document creation device, a first recording unit may record enterprise resource planning implementation procedures and maintenance know-how, and the creation unit may create a development document corresponding to the recorded enterprise resource planning implementation procedures and maintenance know-how as an deliverable. This enables the development document creation device to perform work that is not dependent on the user's ERP expertise or AI skills.

[0071] (Aspect 4) In one embodiment of the development document creation device, the creation unit may create development documents based on a first prompt that is a pre-templated enterprise resource planning implementation work prompt, and a second prompt based on a dialogue with the user. This enables the development document creation device to perform work that is not dependent on the user's ERP expertise or AI skills.

[0072] (Aspect 5) One embodiment of the development document creation device may further include a second recording unit that stores a prompt created by the creation unit and used to create the development document, a shared display unit configured to enable other users within the same team to view and execute a project in which one user is in charge, and a reflection unit that reflects the prompt when creating another development document different from the development document. This enables the development document creation device to perform work that is not dependent on the user's ERP expertise or AI skills.

[0073] (Aspect 6) In the development document creation device of the aspect, the question generator may further generate a second question that complements the first question based on a first user answer from the user. This enables the development document creation device to more accurately create development-specific documents such as in-house proposals, environment definition documents, requirements definition documents, design documents, and plans in system development.

[0074] (Aspect 7) In one embodiment of the development document creation device, the answer acquisition unit may acquire a second user answer to the second question from the user, and the creation unit may create the development document based on the development information, the first answer, and the second answer. This enables the development document creation device to more accurately create development-specific documents such as in-house proposals, environment definition documents, requirements definition documents, design documents, and plans in system development.

[0075] (Aspect 8) In one aspect of the development document creation device, the first prompt may include a conversation history between the user and the AI ​​that is necessary for creating the development document. This enables the development document creation device to more accurately create development-specific documents such as in-house proposals, environment definition documents, requirements definition documents, design documents, and plans in system development.

[0076] (Aspect 9) In one aspect of the development document creation device, the answer generation unit may generate an answer corresponding to a task category. This enables the development document creation device to more accurately create development-specific documents such as in-house proposals, environment definition documents, requirements definition documents, design documents, and plans in system development.

[0077] (Aspect 10) In one aspect of the development document creation device, the answer generation unit may generate an answer according to a current date. This enables the development document creation device to more accurately create development-specific documents such as in-house proposals, environment definition documents, requirements definition documents, design documents, and plans in system development. [Explanation of symbols]

[0078] 1. Introduction process template 2. Workers 3 ERP implementation knowledge and system prompts 4 Navigation and deliverables 5. Execution environment 6 Server configuration environment 7 Project list screen 8. New project creation screen 10. Development Information Acquisition Department 11. Requirements Specification Template 12 Feature List Template 13 Additional Development Basic Design Document 14. Additional Development Programs 20 First Recording Section 30 Task Acquisition Section 31 ERP implementation know-how 32 ERP Standards 33 Document template 34 Web Updates 40 Question generation part 50 Answer acquisition part 51 Project list screen 52 Execution screen 53 Setting screen 54 Mask Keyword Registration 60 Answer generation part 61 Management screen 62 Project template registration 63 Process Registration 64 Task registration (prompt) 65 categories registered 66 Publication Registration 67 License Registration 70 Creation Department 71 ID 72 Name 73 Templates 74 Target Categories 75 Creators 76 Created Date 80 Second Recording Section 81 Templates 82 Confirmation 83 order 84 Process ID 85 Project name 90 Shared display area 91 Project Templates 91-1 Process Template 1 91-2 Process Template 2 91-3 Process Template 3 100 Development document creation device 110 Reflection section 200 User terminals 300 Large-scale language model server 500 Network

Claims

1. A development document creation device that supports the creation of development documents using a large-scale language model, a development information acquisition unit that acquires development information regarding the development described in the development document; a first recording unit that records a prompt of the large-scale language model as template information in response to the development task; a task acquisition unit that acquires the development task from a user; a question generation unit that generates a first question by using the large-scale language model based on template information corresponding to the acquired task; an answer acquisition unit that acquires a first user answer to the first question from a user; an answer generation unit that generates an answer that is a solution based on the first user answer acquired by the answer acquisition unit; a creation unit that creates a development document using the large-scale language model based on the development information and the first user answer to the first question acquired from the user; A development document creation device comprising:

2. The development documents include an in-house planning document, an environment definition document, a requirements definition document, a design document, and a plan document, The creation department creates the in-house production plan, creates a current business flow, a task list, and the environment definition document based on the in-house production plan, creates the requirements definition document based on the current business flow, the task list, and the environment definition document, and creates the design document and the plan based on the created requirements definition document. The development document creation device according to claim 1 .

3. The first recording unit records the introduction procedure and maintenance know-how of enterprise resource planning; The creation unit creates, as a deliverable, the development documentation corresponding to the recorded enterprise resource planning implementation procedure and maintenance know-how. The development document creation device according to claim 1 .

4. the creation unit creates the development document based on a first prompt, which is a template of an introduction work prompt for the enterprise resource planning in advance, and a second prompt based on a dialogue with a user; The development document creation device according to claim 3.

5. a second recording unit that stores a prompt used to create the development document created by the creating unit; A shared display unit configured to allow other users in the same team to view and execute a project that one user is in charge of; a reflection unit that reflects the prompt when creating another development document different from the development document; The development document generation apparatus of claim 1 further comprising:

6. The question generator further generates a second question that complements the first question based on the first user answer from the user. The development document creation device according to claim 1 .

7. the answer acquisition unit acquires a second user answer to the second question from the user; the creation unit creates the development document based on the development information, the first user response, and the second user response. The development document creation device according to claim 6.

8. The first prompt includes a conversation history between the user and the AI ​​necessary to create the development document, 5. The development document creation device according to claim 4.

9. The development document creation device according to claim 7 , wherein the answer generation unit generates the answer in accordance with a category of the task.

10. The development document creation device according to claim 7 , wherein the answer generation unit generates the answer according to a current date.

11. A development documentation creation method for supporting the creation of development documentation using a large-scale language model, comprising: On the computer, a development information acquisition step of acquiring development information regarding the development described in the development document; a recording step of recording a prompt of the large-scale language model as template information according to the development task; a task acquisition step of acquiring a task for the development from a user; a question generation step of generating a first question using the large-scale language model based on the acquired template information corresponding to the task; an answer acquisition step of acquiring a first user answer to the first question from a user; an answer generating step of generating an answer that is a solution based on the first user answer acquired in the answer acquiring step; a creating step of creating a development document using the large-scale language model based on the development information and a first user answer to the first question obtained from the user; How to create development documentation to execute the above.

12. A development documentation creation program that supports the creation of development documentation using a large-scale language model, On the computer, a development information acquisition function for acquiring development information regarding the development described in the development document; a recording function for recording a prompt of the large-scale language model as template information according to the development task; a task acquisition function for acquiring a task of the development from a user; a question generation function that generates a first question using the large-scale language model based on template information corresponding to the acquired task; an answer acquisition function for acquiring a first user answer to the first question from a user; an answer generation function that generates an answer that is a solution based on the first user answer acquired by the answer acquisition function; a creation function of creating a development document using the large-scale language model based on the development information and a first user answer to the first question acquired from the user; A development documentation program that makes this possible.

Images