Information processing device, information processing program, and information processing method

Abstract

To reduce the effort required to create deliverables in system development. [Solution] The information processing device according to the present invention includes a generation unit that inputs a second prompt, which is a prompt that includes an instruction to generate a question for a user who desires the generation of deliverables to improve a first prompt, which is a prompt that is input to a generation AI in order to generate deliverables in system development, and generates a question.

Description

Technical Field

[0005] , , [Figure 5] , [Figure 4] , [Figure 3] , [Figure 2] , [Figure 1]

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

Background Art

[0002] Conventionally, various techniques for supporting developers in system development have been known. For example, a technique for supporting the management of documents (hereinafter, may be referred to as "deliverables") generated in each development process of software development or system development processes is known (Patent Document 1).

Prior Art Documents

Patent Documents

[0003]

Patent Document 1

Summary of the Invention

Means for Solving the Problems

[0004] The information processing apparatus according to the present application includes a generation unit that inputs a second prompt, which is a prompt including an instruction to generate a question for a user who desires to generate the deliverable in order to improve a first prompt that is a prompt input to a generation AI for generating a deliverable in system development, to the generation AI to generate the question.

Brief Description of the Drawings

[0005] [Figure 1] FIG. 1 is a diagram showing an overview of information processing according to an embodiment. [Figure 2] FIG. 2 is a diagram showing an overview of information processing according to an embodiment. [Figure 3] FIG. 3 is a diagram showing a configuration example of an information processing system according to an embodiment. [Figure 4] FIG. 4 is a diagram showing a configuration example of an information processing apparatus according to an embodiment. [Figure 5]Figure 5 shows an example of a prompt generation sheet according to the embodiment. [Figure 6] Figure 6 shows an example of a prompt generation sheet according to the embodiment. [Figure 7] Figure 7 shows an example of a prompt generation sheet according to the embodiment. [Figure 8] Figure 8 is a hardware configuration diagram showing an example of a computer that implements the functions of an information processing device. [Modes for carrying out the invention]

[0006] The following describes in detail, with reference to the drawings, the embodiments for implementing the information processing device, information processing program, and information processing method according to the present application (hereinafter referred to as "embodiments"). Note that these embodiments do not limit the information processing device, information processing program, and information processing method according to the present application. Furthermore, the same parts are denoted by the same reference numerals in each of the following embodiments, and redundant explanations are omitted. Also, in the following, AI (artificial intelligence) and machine learning models are treated as synonymous terms.

[0007] (Embodiment) [1. Introduction] Conventionally, various technologies have been known to support development personnel (e.g., developers) in system development. For example, technologies that support the management of documents (hereinafter sometimes referred to as "deliverables") generated at each development stage of the software development and system development process are known (Patent Document 1). However, conventionally, technologies that support the creation of deliverables in system development have not been sufficient.

[0008] Deliverables in system development include system requirements specifications and business requirements specifications. Traditionally, creating these deliverables required developers to collect, organize, and document a large amount of information necessary for their creation. For example, to create a deliverable for a specific stage of system development (e.g., a system requirements specification), information from deliverables corresponding to downstream stages (e.g., business requirements specifications) is necessary. Therefore, traditionally, developers had to collect, organize, and document information from deliverables corresponding to downstream stages (hereinafter sometimes referred to as "downstream information"). Furthermore, traditionally, the client who requested the system development and the developers had to exchange questions and answers multiple times to collect, organize, and document the information necessary for creating the deliverables. Thus, traditionally, creating deliverables in system development required a great deal of effort.

[0009] In contrast, the information processing device 100 according to the embodiment uses a generation AI to generate questions for a user who desires the generation of deliverables, and repeatedly obtains answers from the user. Specifically, the information processing device 100 inputs a second prompt to the generation AI, which is a prompt that includes an instruction to generate questions for a user who desires the generation of deliverables in order to improve a first prompt, which is a prompt that is input to the generation AI in order to generate deliverables in system development, and generates questions. As a result, the information processing device 100 presents the questions generated by the generation AI to the user and obtains answers from the user. Here, the questions for the user correspond to, for example, questions from the development engineer to the client who requested the development of the system. Also, obtaining answers from the user corresponds to, for example, the development engineer obtaining answers from the client who requested the development of the system. As a result, the information processing device 100 can reduce the effort required to obtain the information necessary to create deliverables. For example, the information processing device 100 can reduce the effort required for the client who requested the development of the system and the development engineer to exchange questions multiple times. Therefore, the information processing device 100 can reduce the effort required to create deliverables in system development.

[0010] Figure 1 is a diagram illustrating an overview of information processing according to the embodiment. Figure 1 is a diagram illustrating an overview of information processing performed by the information processing device 100 according to the embodiment. In Figure 1, the information processing device 100 acquires information (hereinafter sometimes referred to as "downstream information") of deliverables corresponding to processes downstream from a predetermined process deliverable (e.g., a system requirements definition document) desired by the user. The information processing device 100 inputs a first prompt 4, which is a prompt to be input to the generation AI in order to generate deliverables in system development, a question 6 for the user who desires the generation of deliverables to improve the first prompt 4, and a second prompt 2 which includes an instruction to generate a suggestion 5, which is reference information for the user when answering the question 6, and generates the first prompt 4, the suggestion 5, and the question 6.

[0011] Furthermore, if the information processing device 100 generates the first prompt 4, suggestion 5, and question 6, it generates a prompt generation sheet, which is a spreadsheet containing the first prompt 4, suggestion 5, and question 6. If the information processing device 100 generates the prompt generation sheet, it provides the prompt generation sheet to the user's terminal 10. The user views the prompt generation sheet and enters the answer 7 to question 6 into the prompt generation sheet. The user terminal 10 sends the prompt generation sheet with the answer 7 entered to the information processing device 100. The information processing device 100 also retrieves the prompt generation sheet containing the answer 7 from the user's terminal 10. The information processing device 100 also retrieves the answer 7 from the prompt generation sheet. Furthermore, if the information processing device 100 retrieves the answer 7, it generates an updated second prompt containing the first prompt 4, question 6, and answer 7. The information processing device 100 also inputs the updated second prompt into the generation AI to generate an improved first prompt, a new suggestion, and a new question.

[0012] Figure 2 is a diagram illustrating an overview of information processing according to an embodiment. In Figure 2, the information processing device 100 inputs the initial second prompt to the generating AI and generates the first prompt 4-1, the first suggestion 5-1, and the first question 6-1. Here, the initial second prompt may include an instruction that assigns the generating AI the role of prompt designer. The initial second prompt may also include an instruction that instructs the generating AI to generate prompts that are optimal for the user's needs according to predetermined conditions. For example, the initial second prompt may include an instruction that, as a predetermined condition, instructs the generating AI to generate improved first prompts, suggestions, and questions by improving upon the first prompt, using reference information as input information. The initial second prompt may also include an instruction that, as a predetermined condition, instructs the generating AI to check the reference information. The initial second prompt may also include information indicating the user's desire to create desired deliverables as reference information.

[0013] When the information processing device 100 generates the first prompt 4-1, the first suggestion 5-1, and the first question 6-1, it generates a first prompt generation sheet (see, for example, Figure 5 described later) containing the first prompt 4-1, the first suggestion 5-1, and the first question 6-1. When the information processing device 100 generates the first prompt generation sheet, it provides the first prompt generation sheet to the user's terminal 10. The user views the first prompt generation sheet and enters the first answer 7-1 to the first question 6-1 into the first prompt generation sheet. The user terminal 10 sends the first prompt generation sheet with the first answer 7-1 entered to the information processing device 100. The information processing device 100 also retrieves the first prompt generation sheet containing the first answer 7-1 from the user's terminal 10. The information processing device 100 also retrieves the first answer 7-1 from the first prompt generation sheet.

[0014] Furthermore, if the information processing device 100 obtains the first response 7-1, it generates an updated second prompt that includes the first first prompt 4-1, the first question 6-1, and the first response 7-1. Here, the updated second prompt includes the information contained in the initial second prompt. The updated second prompt may also include, as reference information, an instruction statement that instructs the generating AI to generate an improved first prompt, suggestion, and question by improving the first prompt, using the first prompt before improvement (e.g., the first first prompt 4-1), the question previously generated by the generating AI (e.g., the first question 6-1), and the answer to the question previously generated by the generating AI (e.g., the first response 7-1) as input information. The information processing device 100 also inputs the updated second prompt to the generating AI to generate an improved second first prompt 4-2, a second suggestion 5-2, and a second question 6-2.

[0015] Furthermore, if the information processing device 100 generates a second first prompt 4-2, a second proposal 5-2, and a second question 6-2, it generates a second prompt generation sheet (see, for example, Figure 6 described later) that includes the second first prompt 4-2, the second proposal 5-2, and the second question 6-2. Also, if the information processing device 100 generates a second prompt generation sheet, it provides the second prompt generation sheet to the user's user terminal 10. The user views the second prompt generation sheet and enters the second answer 7-2 to the second question 6-2 into the second prompt generation sheet. The user terminal 10 sends the second prompt generation sheet with the second answer 7-2 entered to the information processing device 100. The information processing device 100 also retrieves the second prompt generation sheet, including the second answer 7-2, from the user's user terminal 10. Furthermore, if the information processing device 100 obtains a second answer 7-2, it generates an updated second prompt that includes a second first prompt 4-2, a second question 6-2, and a second answer 7-2. The information processing device 100 also inputs the updated second prompt into the generating AI to generate an improved third first prompt 4-3, a third suggestion 5-3, and a third question 6-3.

[0016] In this way, if the information processing device 100 generates the nth first prompt 4-n (where n is a natural number greater than or equal to 1), the nth proposal 5-n, and the nth question 6-n, it generates the nth prompt generation sheet containing the nth first prompt 4-n, the nth proposal 5-n, and the nth question 6-n. If the information processing device 100 generates the nth prompt generation sheet, it provides the nth prompt generation sheet to the user's terminal 10. The user views the nth prompt generation sheet and inputs the nth answer 7-n to the nth question 6-n into the prompt generation sheet. The user terminal 10 sends the nth prompt generation sheet with the nth answer 7-n entered to the information processing device 100. The information processing device 100 also retrieves the nth prompt generation sheet containing the nth answer 7-n from the user's terminal 10. Furthermore, when the information processing device 100 obtains the nth answer 7-n, it generates an updated second prompt that includes the nth first prompt 4-n, the nth question 6-n, and the nth answer 7-n. The information processing device 100 also inputs the updated second prompt to the generation AI to generate an improved (n+1)th first prompt 4-(n+1), the (n+1)th suggestion 5-(n+1), and the (n+1)th question 6-(n+1). Note that as n increases, the content of the nth first prompt 4-n becomes more specific (see Figure 7 below). Also, as n increases, the content and number of the nth question 6-n converge (see Figure 7 below). The information processing device 100 may terminate the process of generating the prompt generation sheet when n reaches a predetermined threshold (e.g., 10).

[0017] Also, when the information processing apparatus 100 finishes the process of generating the prompt generation sheet, it acquires the nth improved first prompt from the last generated prompt generation sheet. Further, the information processing apparatus 100 inputs the nth first prompt and downstream information into the generation AI to generate a deliverable in system development. Thereby, the information processing apparatus 100 can reduce the labor related to creating the deliverable in system development. Also, the information processing apparatus 100 can generate a higher-quality deliverable based on the first prompt improved to more specific content.

[0018] Also, conventionally, when creating a deliverable in system development, it was necessary to evaluate the created deliverable by, for example, conducting a review by the development responsible person or the like. Thus, conventionally, labor was also required to evaluate the deliverable in system development. In contrast, the information processing apparatus 100 evaluates the deliverable in system development using the generation AI. Thereby, the information processing apparatus 100 can reduce the labor related to evaluating the deliverable in system development.

[0019] Note that the first prompt to the fifth prompt appearing in the specification of this application may be a user prompt. Also, the first prompt to the fifth prompt may be a prompt generated by combining a system prompt and a user prompt. Note that the first prompt to the fifth prompt may be a system prompt.

[0020] The system prompt is a prompt that gives instructions to the AI model regarding formality. Specifically, for example, it may be the language to be used, the attributes of the AI model that the AI model should consider, the processing to be performed on the AI model, the specification of the output format, etc. The language to be used may be, for example, Japanese, English, French, or the like. Also, the attributes of the AI model that the AI model should consider may be, for example, software engineer, programmer, project manager, or system architect, etc. The processing to be performed on the AI model may be, for example, review of the requirements specification, assistance in considering the contents described in the requirements specification, organization of the items described in the requirements specification, etc. And the output format may be, for example, plain text, structured data (JSON format, XML format, yaml format, etc.), unstructured data (pdf, etc.).

[0021] The user prompt is a prompt that gives instructions to the AI model regarding the practical aspect. User prompts may be prepared for each type of processing to be performed on the AI model. Specifically, for example, there may be a user prompt that causes the AI model to detect NG words included in the requirements specification including similar expressions, a user prompt that checks whether all the items to be described in the requirements specification are described without omission, or a user prompt that collects the contents to be described in the requirements specification in an interactive format, etc.

[0022] The execution prompt may be generated by combining the system prompt and the user prompt. For example, it may be a simple combination of the system prompt and the user prompt, and headings such as <user prompt> and <system prompt> may be added to each. The execution prompt can be arbitrarily selected considering the learning content, degree of learning, etc. of the AI model. For example, for a user prompt that causes the AI model to detect NG words included in the requirements specification including similar expressions, for example, the NG words to be detected, an instruction to propose words similar in meaning to the NG words, the number of words to be proposed, the specification of words to be excluded from the similar NG words, whether duplicates are allowed, etc. may be specified. Also, when outputting in structured data, the specific structure, headings, etc. of the data may be specified.

[0023] [2. Configuration of the Information Processing System] Figure 3 shows an example of the configuration of an information processing system 1 according to an embodiment. As shown in Figure 3, the information processing system 1 according to an embodiment includes a user terminal 10 and an information processing device 100. The user terminal 10 and the information processing device 100 are connected to each other via a predetermined communication network (network N) by wired or wireless means.

[0024] The user terminal 10 is an information processing device used by a user, for example, a system development engineer. For example, the user terminal 10 may be an information processing device such as a desktop PC (Personal Computer) or a notebook PC. Alternatively, the user terminal 10 may be a smart device such as a smartphone or tablet. The user terminal 10 displays information received from the information processing device 100, etc., using a web browser or application.

[0025] The information processing device 100 is an information processing device that performs information processing according to the embodiment. The information processing device 100 is implemented by a server device, a cloud system, or the like. For example, the information processing device 100 may perform information processing according to the embodiment in accordance with the information processing method implemented by the information processing program according to the embodiment.

[0026] [3. Configuration of the Information Processing Device] Figure 4 shows an example of the configuration of an information processing device 100 according to the embodiment. The information processing device 100 according to the embodiment includes a communication unit 110, a storage unit 120, and a control unit 130.

[0027] (Communications Department 110) The communication unit 110 is connected to the network N by wire or wireless connection and transmits and receives information with the user terminal 10. For example, the communication unit 110 can be implemented using a NIC (Network Interface Card) or an antenna.

[0028] (Storage unit 120) The memory unit 120 is implemented by, for example, a semiconductor memory element such as RAM (Random Access Memory) or flash memory, or a storage device such as a hard disk or optical disc. Specifically, the memory unit 120 stores the information processing program according to the embodiment. The memory unit 120 also stores information related to the generated AI. For example, the memory unit 120 stores information related to large language models (LLMs) as the generated AI. The generated AI may be provided in the form of SaaS (Software as a Service).

[0029] (Control unit 130) The control unit 130 is a controller, and is realized, for example, by executing various programs stored in the memory device inside the information processing device 100 using RAM as the working area, using a CPU (Central Processing Unit) or MPU (Micro Processing Unit), etc. Alternatively, the control unit 130 is a controller and can be realized, for example, by an integrated circuit such as an ASIC (Application Specific Integrated Circuit) or FPGA (Field Programmable Gate Array).

[0030] The control unit 130 has an acquisition unit 131, a generation unit 132, and a provision unit 133 as functional units, and may realize or execute the information processing operations described below. Note that the internal configuration of the control unit 130 is not limited to the configuration shown in Figure 4, and other configurations are also acceptable as long as they perform the information processing described later. Furthermore, each functional unit represents the function of the control unit 130 and does not necessarily have to be physically separated.

[0031] (Acquisition part 131) The acquisition unit 131 acquires various types of information. The acquisition unit 131 acquires downstream information extracted from downstream deliverables corresponding to processes downstream of the process corresponding to the deliverable in the system development process. The acquisition unit 131 acquires downstream information extracted from deliverables (e.g., business requirements definition document) corresponding to processes downstream of a predetermined process deliverable (e.g., system requirements definition document) desired by the user. Specifically, the acquisition unit 131 acquires downstream deliverables from a deliverable database that stores information about deliverables. For example, the acquisition unit 131 acquires downstream deliverables generated in spreadsheet format. For example, the acquisition unit 131 acquires a spreadsheet which is a business requirements definition document corresponding to a process downstream of the process corresponding to the system requirements definition document. Subsequently, the acquisition unit 131 extracts the text contained in each cell from each cell of the spreadsheet, which is the downstream deliverable, as downstream information. For example, the acquisition unit 131 extracts text from each cell of the spreadsheet which is a business requirements definition document.

[0032] (Generation unit 132) The generation unit 132 generates various information. The generation unit 132 inputs a second prompt (second prompt 2 in Figure 1) to the generation AI, which is a prompt that includes an instruction to generate a question (question 6 in Figure 1) for a user who desires the generation of deliverables to improve the first prompt (first prompt 4 in Figure 1), which is a prompt that is input to the generation AI to generate deliverables in system development, and generates a question to improve the first prompt. Specifically, the generation unit 132 inputs a second prompt (second prompt 2 in Figure 1) that includes an instruction to generate the first prompt (first prompt 4 in Figure 1) to the generation AI to generate the first prompt, and then inputs a second prompt that includes an instruction to generate a question (question 6 in Figure 1) to improve the generated first prompt to the generation AI to generate a question to improve the generated first prompt.

[0033] Furthermore, the generation unit 132 inputs a second prompt (second prompt 2 in Figure 1) to the generation AI, which includes an instruction to generate a suggestion (suggestion 5 in Figure 1), which is reference information for the user when answering a question (question 6 in Figure 1) for improving the first prompt, and generates a suggestion. For example, the generation unit 132 inputs a second prompt to the generation AI, which includes an instruction to generate a question (question 6 in Figure 1), and then inputs a second prompt to the generation AI, which includes an instruction to generate a suggestion (suggestion 5 in Figure 1), which is reference information for the user when answering the generated question, and generates a suggestion.

[0034] Here, generative AI refers to a generative AI that generates text. For example, generative AI is a language model capable of generating text using natural language processing. For example, generative AI is a large language model (LLM). For example, generative AI is a Generative Pre-trained Transformer (GPT).

[0035] For example, the generation unit 132 may input a second prompt to the generation AI that includes an instruction that assigns the generation AI the role of prompt designer, and generate a first prompt, a question, and a suggestion. Alternatively, the generation unit 132 may input a second prompt to the generation AI that includes an instruction that instructs the generation AI to generate a prompt that is optimal for the user's needs according to predetermined conditions, and generate a first prompt, a question, and a suggestion. For example, the generation unit 132 may input a second prompt to the generation AI that includes an instruction that instructs the generation AI to generate an improved first prompt, a suggestion, and a question by improving the first prompt, using reference information as input information for the generation AI, as a predetermined condition, and generate an improved first prompt, a suggestion, and a question. Alternatively, the generation unit 132 may input a second prompt to the generation AI that includes an instruction that instructs the generation AI to check the reference information, as a predetermined condition, and generate a first prompt, a question, and a suggestion. For example, the generation unit 132 may input a second prompt to the generation AI, which includes, as reference information, information indicating the user's desire to create a desired deliverable, and generate the first prompt, question, and suggestion. Alternatively, the generation unit 132 may input a second prompt to the generation AI, which includes, as reference information, the first prompt before improvement, the question previously generated by the generation AI, and the answer to the question previously generated by the generation AI, and generate the first prompt, question, and suggestion. Furthermore, the generation unit 132 may input a second prompt to the generation AI, which includes an instruction to describe the improved first prompt, suggestion, and question in a predetermined structured data format (e.g., JSON format), and generate the first prompt, question, and suggestion. Alternatively, the generation unit 132 may input a second prompt containing downstream information to the generation AI and generate the first prompt, question, and suggestion.

[0036] Furthermore, when the generation unit 132 generates a first prompt, a suggestion, and a question, it generates a prompt generation sheet, which is a spreadsheet containing the first prompt, the suggestion, and the question. Figures 5 to 7 show examples of prompt generation sheets according to the embodiment. Figure 5 shows the first prompt generation sheet SS1 (the first prompt generation sheet described in Figure 2) that is generated for the first time. In Figure 5, the generation unit 132 generates a prompt generation sheet SS1 that includes a cell P11 containing the first prompt, a cell S11 containing the suggestion, a plurality of cells Q11 to Q15 each containing one of the plurality of questions for the first time, and a plurality of cells A11 to A15 for entering the user's answers to each of the plurality of questions for the first time.

[0037] As shown in cell P11 of Figure 5, the generation unit 132 generates the following text as the first prompt: "You want to create a system requirements specification. To do this, you need detailed information such as the system's purpose, functions, performance requirements, user interface, system interface, data management, and constraints. Can you provide this information?" Also, as shown in cell S11 of Figure 5, the generation unit 132 generates the following text as the first suggestion: "To create a system requirements specification, you need specific details about the system. Therefore, we suggest that you change the prompt to a specific question format so that you can provide specific information."

[0038] Furthermore, as shown in cell Q11 of Figure 5, the generation unit 132 generates the text "What is the purpose of the system?" as the first question. In response to this question, as shown in cell A11 of Figure 5, the user enters the text "An e-commerce site for purchasing smartphones." Furthermore, as shown in cell Q12 of Figure 5, the generation unit 132 generates the text "What are the main functions that the system provides?" as the first question. In response to this question, as shown in cell A12 of Figure 5, the user enters text indicating three functions: a smartphone search function, a smartphone comparison function, and a purchase function, along with text indicating an overview and details of each of the three functions. Furthermore, as shown in cell Q13 of Figure 5, the generation unit 132 generates the text "What are the system's performance requirements?" as the first question. In response to this question, as shown in cell A13 of Figure 5, the user enters the text "User screen operations must be completed within 1 second." Furthermore, as shown in cell Q14 of Figure 5, the generation unit 132 generates the text "What is the user interface like?" as the first question. In response to this question, as shown in cell A14 of Figure 5, the user enters the text "We provide a browser-based user interface for PCs and smartphones." Also, as shown in cell Q15 of Figure 5, the generation unit 132 generates the text "What does the system interface look like?" as the first question. In response to this question, as shown in cell A15 of Figure 5, the user enters the text "Payment processing is handled by the payment management system, so payment information is linked to the payment management system when purchasing on a smartphone."

[0039] Figure 6 shows the prompt generation sheet SS2 (the second prompt generation sheet described in Figure 2) that was generated for the second time. In Figure 6, the generation unit 132 generates the prompt generation sheet SS2 which includes cell P21 containing the second first prompt which has been improved from the first first prompt, cell S21 containing the second proposal, multiple cells Q21 to Q26 each containing one of the multiple questions from the second time, and multiple cells A21 to A26 for entering the user's answers to each of the multiple questions.

[0040] As shown in cell P21 of Figure 6, the generation unit 132 generates the following text as the second first prompt: "You want to create a system requirements definition document for an e-commerce site where you purchase smartphones. To do this, you need detailed information such as the system's purpose, functions, performance requirements, user interface, system interface, data management, and constraints. Your system will provide three main functions: searching for, comparing, and purchasing smartphones, and has a performance requirement that user screen operations be processed within one second. The user interface will be provided via a browser for both PCs and smartphones, and data management will be handled by an RDB. There are no particular constraints. Payment processing will be handled by a payment management system, and payment information will be linked to the payment management system when a smartphone is purchased. Based on this information, could you please tell us your requirements in more detail?" In this way, the second first prompt is improved to be more specific than the first first prompt, taking into account the content of the first question and the first answer. Furthermore, as shown in cell S21 of Figure 6, the generation unit 132 generates the following text as a second proposal: "More specific information is needed about the system's functions. For example, what search conditions can be set for the smartphone's search function, and what kind of information can be compared for the comparison function? Having information about the specific operation flow and screen design will allow for a more specific requirements definition."

[0041] Furthermore, as shown in cell Q21 of Figure 6, the generation unit 132 generates the text "What search conditions can be set in the smartphone's search function?" as the second question. In response to this question, as shown in cell A21 of Figure 6, the user enters the text "Manufacturer, model name, price (range), color (multiple selections possible), storage capacity". Furthermore, as shown in cell Q22 of Figure 6, the generation unit 132 generates the text "What information can be compared in the comparison function?" as the second question. In response to this question, as shown in cell A22 of Figure 6, the user enters the text "Manufacturer, CPU, price (range), color (multiple selections possible), storage capacity, resolution, camera performance". Furthermore, as shown in cell Q23 of Figure 6, the generation unit 132 generates the text "What payment methods can be selected in the purchase function?" as the second question. In response to this question, as shown in cell A23 of Figure 6, the user enters the text "Convenience store payment, A company bundled payment, credit card". Furthermore, as shown in cell Q24 of Figure 6, the generation unit 132 generates the text "What are the details of the user interface?" as the second question. In response to this question, as shown in cell A24 of Figure 6, the user enters the text "It is displayed on the screen in HTML." Furthermore, as shown in cell Q25 of Figure 6, the generation unit 132 generates the text "What are the details of the system interface?" as the second question. In response to this question, as shown in cell A25 of Figure 6, the user enters the text "It communicates via HTTPS, and requests and responses are in JSON format." Furthermore, as shown in cell Q26 of Figure 6, the generation unit 132 generates the text "Regarding data management, what kind of data is managed?" as the second question. In response to this question, as shown in cell A26 of Figure 6, the user enters the text "User information, payment information, user access history." Thus, the content of the second question is more specific than the content of the first question, taking into account the second proposal.

[0042] Figure 7 shows the prompt generation sheet SS5 generated for the fifth time. Figure 7 shows how, as n increases, the content of the first prompt becomes more specific, and the number and content of questions converge. In Figure 7, the generation unit 132 generates the prompt generation sheet SS5, which includes cell P51 containing the fifth first prompt which has been improved from the fourth first prompt, cell S51 containing the fifth proposal, multiple cells Q51 to Q53 containing each of the multiple questions of the fifth, and multiple cells A51 to A53 for inputting the user's answers to each of the multiple questions.

[0043] As shown in cell P51 of Figure 7, the generation unit 132, as the fifth first prompt, states: "You want to create a system requirements definition document for an e-commerce site where you can purchase smartphones. To do this, you need detailed information such as the system's purpose, functions, performance requirements, user interface, system interface, data management, and constraints. Your system will provide three main functions: searching, comparing, and purchasing smartphones. The search function allows you to set conditions such as manufacturer, model name, price (range), color (multiple selections possible), storage capacity, and resolution. The comparison function allows you to compare manufacturer, CPU, price (range), color (multiple selections possible), storage capacity, resolution, and camera performance. The purchase function allows you to select convenience store payment, A-company bundled payment, or credit card. The performance requirement is that user screen operations must be completed within 1 second. The user interface will be displayed on the screen in HTML." It will be provided via a browser for PCs and smartphones. The system interface will use HTTPS for communication, and requests and responses will be in JSON format. Data management will be handled by an RDB, which will manage user information, payment information, and user access history. There are no particular constraints. Payment processing will be handled by a payment management system, and payment information will be linked to the payment management system when a smartphone purchase is made. Regarding system operation and maintenance, there are requirements for monitoring the availability of the server, middleware, and application, monitoring error logs in the application logs, and taking daily backups and saving them for one week. As for system security requirements, it is required that an internal vulnerability assessment be completed before release and that critical issues have been resolved. In addition, the system will have a redundant configuration to handle errors and exception situations. Based on this information, could you please tell me your requirements in more detail?" is generated. In this way, the content of the fifth first prompt is considerably more specific than the content of the first prompt of the first and second prompts, taking into account the questions and answers from the first to fourth prompts.Furthermore, as shown in cell S51 of Figure 7, the generation unit 132 generates the following text as the fifth suggestion: "To make the content of the prompt more specific, it would be good to add specific procedures for system operation and maintenance, detailed criteria for security requirements, and specific countermeasures for errors and exceptions."

[0044] Furthermore, as shown in cell Q51 of Figure 7, the generation unit 132 generates the text "What are the specific procedures for system operation and maintenance?" as the fifth question. In response to this question, as shown in cell A51 of Figure 7, the user inputs text stating that the monitoring requirements are server availability monitoring, various MW availability monitoring, and application availability monitoring, and that the maintenance requirements are to consider maintenance and provide a maintenance screen. Also, as shown in cell Q52 of Figure 7, the generation unit 132 generates the text "What are the detailed criteria for security requirements?" as the fifth question. In response to this question, as shown in cell A52 of Figure 7, the user inputs the text "None in particular". Also, as shown in cell Q53 of Figure 7, the generation unit 132 generates the text "What are the specific countermeasures for errors and exception situations?" as the fifth question. In response to this question, as shown in cell A53 of Figure 7, the user inputs the text "None in particular". In this way, the content of the fifth question converges compared to the content of the first and second questions. Furthermore, the number of questions in the fifth round decreases (converges) compared to the number of questions in the first and second rounds.

[0045] (Provider 133) The providing unit 133 provides various information. The providing unit 133 transmits various information to the user's user terminal 10. Specifically, the providing unit 134 provides the prompt generation sheet generated by the generation unit 132 to the user's user terminal 10. For example, if the generation unit 132 generates a prompt generation sheet, the providing unit 134 transmits the prompt generation sheet to the user's user terminal 10. When the user terminal 10 receives the prompt generation sheet, it displays the prompt generation sheet on its screen. The user views the prompt generation sheet displayed on the screen of the user terminal 10 and enters their answers to the questions into the prompt generation sheet. The user terminal 10 also transmits the prompt generation sheet with the answers entered to the questions to the information processing device 100 in response to the user's operation to transmit the prompt generation sheet to the information processing device 100. The acquisition unit 131 acquires the prompt generation sheet, including the answers, from the user's user terminal 10. The acquisition unit 131 also acquires the answers from the prompt generation sheet. In this way, the acquisition unit 131 acquires the user's answers to the questions.

[0046] Furthermore, if the acquisition unit 131 acquires an answer, the generation unit 132 generates an updated second prompt including a first prompt, a question, and an answer. The generation unit 132 also inputs the updated second prompt to the generation AI to generate an improved first prompt, a suggestion, and a question. For example, the generation unit 132 inputs a second prompt to the generation AI along with the answer, which includes an instruction to generate an improved first prompt based on the content of the answer, thereby generating an improved first prompt.

[0047] Furthermore, the generation unit 132 inputs the improved first prompt into the generation AI to generate the deliverable. For example, the generation unit 132 determines whether the number of prompt generation sheets has reached a predetermined threshold (e.g., 10 times). If the generation unit 132 determines that the number of prompt generation sheets has reached the predetermined threshold, it obtains the improved first prompt from the last generated prompt generation sheet. Also, if the generation unit 132 obtains the improved first prompt, it inputs the improved first prompt into the generation AI to generate the deliverable.

[0048] Furthermore, the generation unit 132 inputs downstream information into the generation AI to generate the deliverable. For example, the generation unit 132 inputs the downstream information into the generation AI as input information for the generation AI, along with an instruction statement that instructs the generation AI to generate the deliverable, and generates the deliverable. The generation unit 132 inputs the improved first prompt and downstream information into the generation AI to generate the deliverable.

[0049] Furthermore, the generation unit 132 generates deliverables described in a predetermined structured data format (e.g., JSON format). For example, the generation unit 132 inputs a prompt to the generation AI instructing it to generate deliverables described in a predetermined structured data format, and then generates deliverables described in a predetermined structured data format.

[0050] Furthermore, the generation unit 132 inputs a third prompt to the generation AI, which is a prompt that includes an instruction to obtain synonyms of the NG word in the deliverable, and obtains synonyms of the NG word. For example, the generation unit 132 inputs a third prompt to the generation AI that instructs it to generate a list of synonyms of the expression "NG word," and obtains a list of synonyms of the NG word. Furthermore, the generation unit 132 inputs a fourth prompt to the generation AI, which is a prompt that includes an instruction to obtain a determination result on whether or not the deliverable contains the NG word and its synonyms, and obtains a determination result on whether or not the deliverable contains the NG word and its synonyms. For example, the generation unit 132 inputs a fourth prompt that instructs it to generate a determination result on whether or not the deliverable contains the expression "NG word" and words included in the list of synonyms of the NG word, and the deliverable, and obtains a determination result on whether or not the deliverable contains the NG word and its synonyms.

[0051] Furthermore, the generation unit 132 inputs a fifth prompt to the generation AI, which is a prompt that includes an instruction to obtain a determination result on whether or not each of the multiple specified items that should be included in the deliverable is included in the deliverable, and obtains a determination result on whether or not each of the multiple specified items is included in the deliverable. For example, the generation unit 132 inputs a fifth prompt to the generation AI that instructs it to generate a list of multiple specified items and a determination result on whether or not each of the multiple specified items is included in the deliverable, and obtains a determination result on whether or not each of the multiple specified items is included in the deliverable.

[0052] [4. Effects] As described above, the information processing device 100 according to the embodiment includes a generation unit 132. The generation unit 132 inputs a second prompt, which is a prompt that includes an instruction to generate a question for a user who desires the generation of deliverables to improve a first prompt, which is a prompt that is input to the generation AI in order to generate deliverables in system development, and generates a question.

[0053] As a result, the information processing device 100 can present questions generated by the generation AI to the user and obtain answers to those questions from the user. Here, the questions to the user correspond to, for example, questions asked by the development engineer to the client who requested the system development. Similarly, obtaining answers from the user corresponds to, for example, the development engineer obtaining answers to questions from the client who requested the system development. This allows the information processing device 100 to reduce the effort required to obtain the information necessary to create deliverables. For example, the information processing device 100 can reduce the effort required for the client who requested the system development and the development engineer to engage in multiple question-and-answer sessions. Therefore, the information processing device 100 can reduce the effort required to create deliverables in system development. Furthermore, because the information processing device 100 can reduce the effort required to create deliverables in system development, it can contribute to achieving Sustainable Development Goal (SDG) 9, "Build resilient infrastructure, promote inclusive and sustainable industrialization and foster innovation."

[0054] Furthermore, the generation unit 132 inputs a second prompt containing an instruction to generate a first prompt to the generation AI to generate a first prompt, and inputs a second prompt containing an instruction to generate a question for improving the generated first prompt to the generation AI to generate a question for improving the generated first prompt.

[0055] As a result, the information processing device 100 generates prompts to be input to the generation AI in order to produce deliverables in system development, presents questions to the user to improve the generated prompts, and obtains answers to the questions from the user.

[0056] The information processing device 100 also includes an acquisition unit 131. The acquisition unit 131 acquires the user's answer to a question. The generation unit 132 inputs a second prompt to the generation AI, along with the answer, which includes an instruction to generate an improved first prompt based on the content of the answer, and generates the improved first prompt.

[0057] This allows the information processing device 100 to obtain answers to questions from the user. Furthermore, the information processing device 100 can generate improved prompts based on the answers obtained from the user, thereby producing higher quality deliverables.

[0058] Furthermore, the generation unit 132 inputs a second prompt to the generation AI, which includes an instruction to generate suggestions that serve as reference information for the user when answering a question, and generates suggestions.

[0059] This allows the information processing device 100 to help the user input an appropriate answer to a question.

[0060] Furthermore, the generation unit 132 inputs the improved first prompt to the generation AI to generate the output.

[0061] As a result, the information processing device 100 can generate higher quality deliverables based on improved prompts with more specific content.

[0062] Furthermore, the generation unit 132 generates deliverables described in a predetermined structured data format.

[0063] As a result, the information processing device 100 can generate deliverables described in a predetermined structured data format.

[0064] Furthermore, the acquisition unit 131 acquires downstream information extracted from downstream deliverables corresponding to processes downstream of the deliverables in the system development process. The generation unit 132 inputs the downstream information into the generation AI to generate deliverables.

[0065] This allows the information processing device 100 to reduce the effort required to collect downstream information.

[0066] Furthermore, the generation unit 132 inputs a third prompt to the generation AI, which is a prompt that includes an instruction to obtain synonyms for the NG word in the deliverable, to obtain synonyms for the NG word, and then inputs a fourth prompt to the generation AI, which is a prompt that includes an instruction to obtain a determination result on whether or not the deliverable contains the NG word and its synonym, to obtain a determination result on whether or not the deliverable contains the NG word and its synonym.

[0067] This reduces the effort required for the information processing device 100 to evaluate whether or not the output contains prohibited words and synonyms.

[0068] Furthermore, the generation unit 132 inputs a fifth prompt to the generation AI, which is a prompt that includes an instruction to obtain a determination result on whether or not each of the multiple specified items that should be included in the deliverable is included in the deliverable, and obtains a determination result on whether or not each of the multiple specified items is included in the deliverable.

[0069] This reduces the effort required for the information processing device 100 to evaluate whether each of the multiple specified items that should be included in the deliverable is included in the deliverable.

[0070] [5. Hardware Configuration] Furthermore, the information processing device 100 according to the above-described embodiment is realized by a computer 1000 having a configuration such as that shown in Figure 8. Figure 8 is a hardware configuration diagram showing an example of a computer that realizes the functions of the information processing device 100. The computer 1000 includes a CPU 1100, RAM 1200, ROM 1300, HDD 1400, communication interface (I / F) 1500, input / output interface (I / F) 1600, and media interface (I / F) 1700.

[0071] The CPU 1100 operates based on programs stored in the ROM 1300 or HDD 1400, controlling various components. The ROM 1300 stores boot programs executed by the CPU 1100 when the computer 1000 starts up, as well as programs that depend on the computer 1000's hardware.

[0072] The HDD1400 stores programs executed by the CPU1100, as well as data used by such programs. The communication interface1500 receives data from other devices via a predetermined communication network and sends it to the CPU1100, and transmits data generated by the CPU1100 to other devices via the predetermined communication network.

[0073] The CPU 1100 controls output devices such as displays and printers, and input devices such as keyboards and mice, via the input / output interface 1600. The CPU 1100 acquires data from input devices via the input / output interface 1600. The CPU 1100 also outputs the generated data to output devices via the input / output interface 1600.

[0074] The media interface 1700 reads a program or data stored in the recording medium 1800 and provides it to the CPU 1100 via the RAM 1200. The CPU 1100 loads the program from the recording medium 1800 onto the RAM 1200 via the media interface 1700 and executes the loaded program. The recording medium 1800 is, for example, an optical recording medium such as a DVD (Digital Versatile Disc) or PD (Phase Change Rewritable Disk), a magneto-optical recording medium such as an MO (Magneto-Optical disk), a tape medium, a magnetic recording medium, or a semiconductor memory.

[0075] For example, when the computer 1000 functions as an information processing device 100 according to the embodiment, the CPU 1100 of the computer 1000 realizes the functions of the control unit 130 by executing a program loaded on the RAM 1200. The CPU 1100 of the computer 1000 reads and executes these programs from the recording medium 1800, but as another example, these programs may be obtained from other devices via a predetermined communication network.

[0076] Although some embodiments of the present invention have been described in detail above with reference to the drawings, these are illustrative examples, and the present invention can be implemented in various other forms with modifications and improvements based on the knowledge of those skilled in the art, starting with the embodiments described in the disclosure section of the invention.

[0077] [6. Other] Furthermore, among the processes described in the above embodiments and modifications, all or part of the processes described as being performed automatically can be performed manually, or all or part of the processes described as being performed manually can be performed automatically by known methods. In addition, the processing procedures, specific names, and information including various data and parameters shown in the above document and drawings can be changed at will unless otherwise specified. For example, the various information shown in each figure is not limited to the information shown.

[0078] Furthermore, the components of each illustrated device are functionally conceptual and do not necessarily need to be physically configured as shown. In other words, the specific forms of distribution and integration of each device are not limited to those shown, and all or part of them can be functionally or physically distributed and integrated in any unit according to various loads and usage conditions.

[0079] Furthermore, the embodiments and modifications described above can be combined as appropriate, provided that the processing content is not inconsistent.

[0080] Furthermore, this disclosure may be implemented, for example, as a software program that runs on a personal computer or an application that runs on a smartphone. Also, the methods and programs related to this disclosure may be provided in SaaS format. [Explanation of Symbols]

[0081] 100 Information Processing Devices 110 Communications Department 120 Storage section 130 Control Unit 131 Acquisition Department 132 Generation part 133 Provision Department

Claims

1. A generation unit inputs a second prompt, which is a prompt that includes instructions to generate a question for a user who desires the generation of deliverables, to the generation AI in order to improve the first prompt, which is a prompt that is input to the generation AI in order to generate deliverables in system development, and generates the question. An information processing device equipped with the following features.

2. The generating unit is Inputting the second prompt, which includes an instruction to generate the first prompt, to the generating AI to generate the first prompt, and inputting the second prompt, which includes an instruction to generate the question for improving the generated first prompt, to the generating AI to generate the question for improving the generated first prompt, The information processing apparatus according to claim 1.

3. The system includes an acquisition unit that obtains the user's response to the aforementioned question, The generating unit is Along with the aforementioned answer, the second prompt, which includes an instruction to generate an improved first prompt based on the content of the aforementioned answer, is input to the generating AI to generate the improved first prompt. The information processing apparatus according to claim 1.

4. The generating unit is The second prompt, which includes an instruction to generate suggestions that serve as reference information for the user to answer the aforementioned question, is input to the generating AI to generate the suggestions. The information processing apparatus according to claim 1.

5. The generating unit is The improved first prompt is input to the generating AI to generate the output. The information processing apparatus according to claim 3.

6. The generating unit is The system generates the deliverable described in a predetermined structured data format. The information processing apparatus according to claim 5.

7. The acquisition unit is, Of the processes in the aforementioned system development, downstream information is extracted from downstream deliverables corresponding to processes downstream of the process corresponding to the aforementioned deliverable. The generating unit is The downstream information is input to the generating AI to generate the output. The information processing apparatus according to claim 5.

8. The generating unit is A third prompt, which is a prompt that includes an instruction to obtain synonyms for the NG word in the deliverable, is input to the generating AI to obtain the synonyms for the NG word, and a fourth prompt, which is a prompt that includes an instruction to obtain a determination result of whether or not the NG word and the synonyms are included in the deliverable, is input to the generating AI to obtain a determination result of whether or not the NG word and the synonyms are included in the deliverable. The information processing apparatus according to claim 5.

9. The generating unit is A fifth prompt, which is a prompt that includes an instruction to obtain a determination result of whether or not each of the multiple specified items that should be included in the deliverable is included in the deliverable, is input to the generating AI to obtain a determination result of whether or not each of the multiple specified items is included in the deliverable. The information processing apparatus according to claim 5.

10. A generation procedure that involves inputting a second prompt, which is a prompt that includes instructions to generate a question for a user who desires the generation of a deliverable, into the generation AI in order to improve the first prompt, which is a prompt that is input into the generation AI in order to generate a deliverable in system development, and generating the question. An information processing program that causes a computer to execute something.

11. An information processing method implemented by a program executed by an information processing device, A generation process that involves inputting a second prompt, which is a prompt that includes instructions to generate a question for a user who desires the generation of a deliverable, into the generation AI in order to improve the first prompt, which is a prompt that is input into the generation AI in order to generate a deliverable in system development, and generating the question. Information processing methods including

Images