JP7799020B1 - Information processing method, program, and information processing system - Google Patents

Abstract

An information processing method, program, and information processing system are provided that support system development by identifying requirements that are not limited in scope of application.
[Solution] In an information processing system, an information processing method executed by an information processing device (server) includes a question providing step S102 that performs at least one of selecting a question and generating a question to obtain requirements, an answer obtaining step S108 that obtains an answer to the question, and output generating steps S110 and S112 that learn at least text data, input information including the answer, and generate requirements inferred from the answer using a trained model to generate the requirements from the answer.
[Selected Figure] Figure 4

Description

The present invention relates to an information processing method, a program, and an information processing system.

Patent Document 1 discloses a requirements definition determination support system that can significantly reduce the number of requirements overlooked during system development. This system aims to prevent oversights in requirements definition by recording items according to a requirements analysis template and making it possible to record and view the contents of question and answer sessions.

Japanese Patent Application Laid-Open No. 2017-049697

However, in the case of the above prior art, since items are recorded according to a template, the content may be limited to the scope assumed by the template, and there is room for improvement in this regard.

The present invention takes the above into consideration and aims to support system development by identifying requirements that are not limited in scope.

According to one embodiment, the information processing method is executed by an information processing device and includes a question providing step of selecting a question to acquire requirements and/or generating the question, an answer acquisition step of acquiring an answer to the question, and an output generation step of generating the requirements from the answers using a trained model that learns at least text data and generates requirements estimated from the answers when information including the answers is input.

According to one embodiment of the program, an information processing device is caused to execute an information processing method including: a question providing step for performing at least one of selecting a question to obtain requirements and generating the question; an answer obtaining step for obtaining an answer to the question; and an output generating step for generating the requirements from the answer using a trained model that learns at least text data and generates requirements estimated from the answer when information including the answer is input.

According to one embodiment, the information processing system is executed by an information processing device and includes a question providing unit that performs at least one of selecting a question to acquire requirements and generating the question, an answer acquisition unit that acquires answers to the questions, and an output generation unit that generates the requirements from the answers using a trained model that learns at least text data and generates requirements inferred from the answers when information including the answers is input.

According to one embodiment, system development can be supported by identifying requirements that are not limited in scope.

FIG. 1 is a diagram illustrating an example of a configuration of an information processing system according to a first embodiment. FIG. 2 is a diagram illustrating an example of a hardware configuration of a server according to the first embodiment. FIG. 2 is a diagram illustrating an example of a functional configuration of a server according to the first embodiment. FIG. 2 is a diagram illustrating an example of a processing flow of the information processing system according to the first embodiment. FIG. 2 is a diagram showing an example of a screen display in the information processing system according to the first embodiment. FIG. 2 is a diagram showing an example of a screen display in the information processing system according to the first embodiment. FIG. 2 is a diagram showing an example of a screen display in the information processing system according to the first embodiment. FIG. 10 is a diagram illustrating an example of a functional configuration of a server according to a second embodiment. FIG. 10 is a diagram illustrating an example of a processing flow of an information processing system according to a second embodiment. FIG. 10 is a diagram showing an example of a screen display in an information processing system according to the second embodiment. FIG. 10 is a diagram conceptually illustrating part of the processing of an information processing system according to a second embodiment. FIG. 11 is a diagram illustrating an example of a functional configuration of a server according to the third embodiment. FIG. 11 is a diagram illustrating a part of an example of a processing flow of an information processing system according to a third embodiment. FIG. 11 is a diagram illustrating another part of the example of the processing flow of the information processing system according to the third embodiment. FIG. 11 is a diagram conceptually illustrating part of the processing of an information processing system according to a third embodiment.

(First embodiment)
A first embodiment of an information processing system according to the present invention will be described below with reference to Figures 1 to 7. Note that the same or equivalent components and parts in each figure are denoted by the same reference numerals. Also, the dimensional proportions in the drawings have been exaggerated for the sake of explanation, and may differ from the actual proportions.

(System overview)
First, an overview of the information processing system 10 according to this embodiment will be described. The information processing system 10 according to this embodiment is a system in which, when a prompt serving as processing instruction information for instructing processing is input to a trained model such as a large-scale language model, the trained model generates a processing result in response to the prompt and outputs the processing result. That is, when a user U inputs a prompt indicating the creation of a requirements definition as a requirement (see the prompt PT in FIG. 5 ), the information processing system 10 performs processing using a trained model based on the prompt and generates a processing result. Furthermore, the information processing system 10 provides questions in response to the prompt and generates a requirements definition from the user U's answers to the questions. Note that the above-described requirements definition is intended to be required for computer-based system development and improvement, but is not limited thereto and may also be required for other purposes, such as the introduction of business processes, the establishment or modification of regulations and compliance. That is, the requirements definition in this embodiment organizes and clarifies the requirements for the system to be developed. Note that the prompt PT is not limited to being directly input by the user U, but may also be automatically generated by the information processing system 10 or based on a pre-prepared template.

(System configuration)
Fig. 1 is a diagram showing an example of the configuration of an information processing system 10 according to this embodiment. As shown in Fig. 1, the information processing system 10 according to this embodiment includes a server 12 as an information processing device and a plurality of user terminals 14, which are communicably connected to each other via a network N. The network N is, for example, a wired local area network (LAN), a wireless LAN, the Internet, a public line network, a mobile data communication network, or a combination thereof.

The user terminal 14 is an example of an information processing device on which the user U, manager M, and operator OP each perform operations to input and display various information. The user terminal 14 may be a PC (Personal Computer), smartphone, tablet terminal, server device, microcomputer, wearable device, or a combination of these. As an example, the user U is a general employee, the manager M is a manager, and the operator OP is a service provider that provides services related to the information processing system 10. The user U, manager M, and operator OP each have account information as user information linked to specific authority information, and by logging in to the information processing system 10 using the account information, the information processing system 10 can obtain the account information and information regarding the usage status of the account information (details will be described later).

The server 12 is an example of an information processing device that acquires information input from the user terminal 14, processes the information, and outputs the results. The server 12 may be a PC (Personal Computer), a smartphone, a tablet terminal, a server device, a microcomputer, or a combination of these. The specific configuration and operation of the server 12 will be described later.

(Hardware configuration)
2 is a block diagram showing the hardware configuration of the server 12. The server 12 includes a processor 120, a memory 122, a storage 124, a communication I/F 126, an input/output I/F 128, an output device 130, an input device 132, and a drive device 134, which are communicatively connected to each other via a bus B.

The processor 120 controls each component of the server 12 and realizes the functions of the server 12 by expanding various programs stored in the storage 124 into the memory 122 and executing them. The programs executed by the processor 120 include, but are not limited to, an OS (Operating System) and the program 220 described below. The execution of these programs by the processor 120 realizes part of the state visualization method according to this embodiment. The processor 120 is, for example, a CPU (Central Processing Unit), an MPU (Micro Processing Unit), a GPU (Graphics Processing Unit), an ASIC (Application Specific Integrated Circuit), a DSP (Digital Signal Processor), or a combination thereof.

Memory 122 is, for example, ROM (Read Only Memory), RAM (Random Access Memory), or a combination thereof. ROM is, for example, PROM (Programmable ROM), EPROM (Erasable Programmable ROM), EEPROM (Electrically Erasable Programmable ROM), or a combination thereof. RAM is, for example, DRAM (Dynamic RAM), SRAM (Static RAM), MRAM (Magnetoresistive RAM), or a combination thereof.

Storage 124 stores the OS, various programs described below, and various data. Storage 124 is, for example, a flash memory, HDD (Hard Disk Drive), SSD (Solid State Drive), SCM (Storage Class Memories), or a combination of these.

The communication I/F 126 is an interface that connects the server 12 to external devices, including the user terminal 14 and the image capture device 16, via the network N and controls communications. The communication I/F 126 is, for example, an adapter that complies with Bluetooth (registered trademark), Wi-Fi (registered trademark), ZigBee (registered trademark), Ethernet (registered trademark), or optical communications (e.g., Fibre Channel), but is not limited to these.

The input/output I/F 128 is an interface for connecting an input device 132 and an output device 130 to the server 12. The input device 132 is, for example, a mouse, keyboard, touch panel, microphone, scanner, camera, various sensors, operation buttons, or a combination of these. The output device 130 is, for example, a display, projector, printer, speaker, vibrator, or a combination of these.

The drive device 134 reads and writes data from the disk media 136. The drive device 134 is, for example, a magnetic disk drive, an optical disk drive, a magneto-optical disk drive, or a combination of these. The disk media 109 is, for example, a CD (Compact Disc), a DVD (Digital Versatile Disc), a FD (Floppy Disk), an MO (Magneto-Optical Disc), a BD (Blu-ray (registered trademark) Disc), or a combination of these.

In this embodiment, the program may be written to memory 122 or storage 124 during the manufacturing stage of server 12, or may be provided to server 12 via network N, or may be provided to server 12 via a non-transitory computer-readable recording medium such as disk media 136.

Furthermore, the hardware configuration of the user terminal 14 is substantially the same as the hardware configuration of the server 12 described above, so a detailed explanation will be omitted.

(Functional configuration)
Next, the functional configuration of the server 12 will be described. Fig. 3 is a diagram showing an example of the functional configuration of the server 12. When executing various programs, the server 12 uses the above-mentioned hardware resources to realize various functions. The server 12 has a communication unit 20, a storage unit 22, and a control unit 24 as the functional configuration realized by the server 12. Each functional configuration is realized when the processor 120 reads and executes a program 220 stored in the memory 122 or the storage 124.

The communication unit 20 is realized by the communication I/F 126. The communication unit 20 sends and receives information to and from the user terminal 14 via the network N. The communication unit 20 receives information input from the user terminal 14. The communication unit 20 also sends information to the user terminal 14 and receives requests from the user U from the user terminal 14.

The memory unit 22 is realized by the memory 122 and storage 124. The memory unit 22 stores a program 220, a trained model 222, an acquired information DB 224, a user information DB 226, a question DB 228, and a processing result DB 230.

The trained model 222 is composed of at least one trained machine learning model. At least one of the machine learning models that make up this trained model 222 is, as an example, a large-scale language model that is trained with large amounts of text data from internet articles, books, websites, etc. When text data is input as a prompt, it generates and outputs text data as a processing result in response to the prompt. The large-scale language model in the trained model 222 of this embodiment generates questions in response to the prompt and generates a requirements definition, requirements definition document, requirements definition, and requirements definition document (details will be described later) that are inferred based on the answers to the questions.

Furthermore, the trained model 222 has the function of generating an evaluation of the answers obtained from user U based on predetermined evaluation criteria and the basis for that evaluation. As a specific example, it analyzes the text data of the answer and evaluates it against evaluation criteria such as the answer's specificity, appropriateness, consistency, and comprehensiveness. For example, it determines whether the answer contains specific numerical values or examples, whether it is technically and logically valid, whether it is consistent with previous answers and requirement definitions, and whether it covers all the information necessary for the question. These evaluation results and the basis for them are provided as feedback to user U, and can encourage them to revise the answer or provide additional information.

The trained model 222 also analyzes the generated requirements definition and detects missing information in at least one of the questions and answers. Specifically, it can analyze the content of the requirements in detail and identify areas where information or details are lacking. If missing information is detected, it generates a prompt to obtain the missing information and the information needed to resolve it, and this is used to create additional questions or generate improved examples of answers (see the improved example IE in Figure 6 as an example) in the trained model 222.

Furthermore, the trained model 222 has the function of determining whether there are any inconsistencies between the multiple generated requirement definitions. It analyzes whether there are any contradictions or inconsistencies between the requirement definitions, and if an inconsistency is detected, it clarifies the details. If it is determined that an inconsistency exists, it generates a prompt to obtain the information necessary to resolve the inconsistency, and this is used when the question providing unit 242 provides an additional question to the user U.

Acquired information DB224 is a database that stores information indicating prompts and answers entered by user U. The prompts and answers are written in natural language, and, as an example, acquired information DB224 stores the prompts and answers in association with the account information used when the prompts and answers were entered.

User information DB226 is a database that stores various information about users U who have registered to use information processing system 10. Examples of the various information stored include user identification information for unique identification, user preference information such as interest information, user history information including usage status such as prompt input information and answer input information, and user attribute information such as occupation, position, and authority.

The user information DB is a database that stores questions presented to user U to obtain a requirement definition. Questions are written in natural language, and the user information DB stores, for example, the question text linked to the account information used when the question was submitted. Specific examples of these questions are categorized as follows:

First, there are questions about the purpose of the system. These questions aim to clarify what the main purpose of the new system is, what business goals the company wants to achieve by implementing the system, and what current challenges and issues the company faces.

Next are questions about users. These questions are used to understand who the target users are, what their technical skill level is, and what features and services they expect from the system.

Functional requirements questions are also included to ascertain what key features the system must have, what the priority of each feature is, and whether there are any features being considered for addition in the future.

In addition, there are questions about non-functional requirements. These questions are stored to ascertain the system's performance requirements (e.g., response time, processing speed), availability and reliability requirements, and security and privacy requirements.

Data questions include questions to clarify what type and format of data will be handled, what data retention periods and archiving requirements are in place, and what requirements there are for backing up and restoring data.

Questions about the user interface include questions to confirm whether there are any requests regarding the user interface design, what are the requirements for multilingual support and accessibility, and whether use on mobile devices is anticipated.

Questions regarding integration with other systems include questions to clarify whether integration with other systems or services is necessary, what the requirements are for APIs and data sharing, and whether there are any integration requirements regarding authentication and authorization.

Questions regarding technical constraints include whether there are any specific platforms, languages, or frameworks to be used, what the technical constraints and prerequisites are for building the system, and whether it is necessary to utilize existing hardware or software assets.

Project management questions include questions to clarify whether there are development schedule or delivery deadline requirements, what budget and resource constraints there are, and who the stakeholders and parties are.

Questions about laws, regulations and standards include questions to confirm whether there are any laws or industry standards that must be complied with, what regulations relate to data protection and privacy, and what compliance requirements there are.

Questions regarding operation and maintenance include questions to clarify what the system's operating hours and maintenance requirements are, what the requirements are for responding to failures and redundancy, and whether there are plans for regular maintenance and updates.

Finally, there are other questions. These are questions to confirm whether there are any risks or issues that require particular attention, what are the requests regarding future business development and scalability, and what information should be communicated to the development team. Through the above questions, the user information DB can systematically collect information necessary for defining system development requirements from user U. This makes it possible to clarify the requirements definition and streamline the system development process. In addition to the specific questions mentioned above, the user information DB may also contain a variety of questions necessary for defining system development requirements.

The processing result DB 230 is a database that stores the results of various processes executed by the server 12, such as responses obtained from the user U, generated requirement definitions, and their analysis results. Specifically, the database stores requirement definitions generated using the trained model 222 in the output generation step, evaluation results and their rationales obtained through evaluation and rationale generation, information regarding information deficiencies and inconsistencies detected in the verification step, and additional questions and their answers (details will be described later). The processing result DB 230 also records and associates each processing result with the user U's account information and timestamps at the time the result was obtained. This enables tracking and analysis of past processing results, contributing to updating requirement definitions and optimizing the system development process. Furthermore, the information stored in the processing result DB 230 is used as basic data for judgments and processing in other processing steps, such as determining dependency relationships and inconsistencies between requirement definitions and setting reference requirement definitions. It also functions as an information source for facilitating interactions with the user U, such as issuing instructions to delete requirement definitions or generating questions to confirm updates.

The control unit 24 is realized by the processor 120 reading and executing the program 220 from the memory 122 (see Figure 2) and working in cooperation with other hardware components. The control unit 24 includes a question providing unit 242, an information acquiring unit 244 as an answer acquiring unit, an information processing unit 246 as an output generating unit, and an output unit 248.

The question providing unit 242 provides a question to the user U (see question Q in Figure 5 as an example of a question). Specifically, the question providing unit 242 at least selects and generates questions. That is, when selecting a question, appropriate questions for defining requirements in system development are selected from the user information DB stored in the memory unit 22. In selecting this question, as a specific example, the question providing unit 242 references the user U's account information stored in the user information DB 226, the input prompt, the history in the acquired information DB 224, etc., to grasp at least one of the instructions and attributes from the user U (e.g., job title, field of expertise, past response history). Next, the grasped instructions and/or attributes are used to clarify issues in the current requirements definition and the need for additional information. Based on this information, the multiple questions stored in the user information DB are prioritized based on the importance and urgency of the information to be collected. Specifically, questions related to items with insufficient information or items in which inconsistencies have been detected are set to a high priority. It is also possible to exclude questions that have already been answered by user U in the past, and include questions that have not yet been answered or questions that require reconfirmation as selection targets.

When selecting questions, the question provider 242 selects appropriate questions for defining requirements in system development from the user information DB stored in the storage unit 22. In this process, the question provider 242 references user U's account information stored in the user information DB 226, the input prompt, and the history in the acquired information DB 224 to understand user U's instructions and attributes (e.g., job title, field of expertise, past response history). Next, based on this information, the provider clarifies the issues and need for additional information in the current requirements definition, and prioritizes the multiple questions stored in the user information DB based on the importance and urgency of the information to be collected, selecting from the top. Specifically, questions related to items with insufficient information or items for which inconsistencies have been detected are set to a high priority. Furthermore, questions that user U has already answered in the past are excluded, and only questions that have not yet been answered or questions requiring reconfirmation are selected. Note that in this embodiment, the selection process is performed on multiple questions stored in the user information DB, but this is not limited to this; the selection process may also be performed on other information sources, such as the processing result DB 230.

When generating questions, the question provider 242 uses the trained model 222 to generate new questions. It clarifies issues with the current requirements definition and the need for additional information based on instructions from user U, past response history, attributes, etc., and automatically inputs analytical instructions (prompts) to the trained model 222 based on this information to analyze user U's needs, information deficiencies, inconsistencies, etc. Based on the analysis results, it identifies new information that should be obtained from user U and generates appropriate questions to elicit that information. The generated questions are optimized for expression according to user U's field of expertise and level of understanding, and are created in a format that is easy for user U to answer, such as by using technical terms and presenting specific examples. Furthermore, in order to deepen requirements definitions in system development, it is also possible to generate questions that resolve relevance and inconsistencies with existing requirements definitions. Furthermore, in response to new responses and reactions from user U, it is possible to dynamically generate questions using the trained model 222 to collect detailed information.

The question providing unit 242 selects and generates questions depending on various factors such as the user U's situation, the state of the requirement definition, and missing or contradictory information, thereby achieving efficient information collection and clarification of the requirement definition. That is, the information processing unit 246 analyzes the requirement definition output in the previous process, and if it detects a lack of information in at least one of the questions and answers, it selects and generates an additional question for user U to acquire the missing information. For example, consider a case where user U presents an "inventory management system" as the system's main function as an answer, but the details are unclear. When the information processing unit 246 analyzes this answer, it detects that specific information such as the type and quantity of products to be managed, the need for real-time inventory updates, and access permissions to inventory data is missing. In this case, in order to acquire the missing information, the question providing unit 242 selects or generates additional questions such as "Please tell me specifically the types and numbers of products managed by the inventory management system," "Does inventory information need to be updated in real time," and "Do access permissions to inventory data differ for each user?" and performs processing to provide them to user U.

As another example, there may be a lack of information regarding non-functional requirements. For example, suppose user U has not provided specific information regarding the performance requirements of the system. If the information processing unit 246 detects this lack, the question providing unit 242 selects or generates questions such as "Is there a target value for the system's response time? For example, the time from user operation to screen display," "What is the maximum expected number of users who will use the system simultaneously?" and "If a large amount of data processing occurs, what level of processing speed is required?" and performs processing to provide these questions to user U, thereby requesting user U to provide detailed information.

As another example, the details of security requirements may be unclear. When user U only requests "high security," the specific requirements for security measures are unclear. When the information processing unit 246 detects this lack of information, the question providing unit 242 selects or generates specific questions, such as "What method of user authentication do you desire (e.g., ID/password, two-factor authentication, biometric authentication, etc.)?", "Is data encryption necessary? If so, which will be encrypted, stored data or communication data?", and "Will security policies be implemented, such as account locking based on the number of failed login attempts or periodic password changes?", and performs processing to provide these to user U, thereby collecting necessary information from user U. Note that, as an example, in this embodiment, when the information processing unit 246 analyzes and detects a lack of information in at least one of the questions and answers, the question providing unit 242 inputs the details of the lack and a prompt that instructs the output of a question to acquire information necessary to resolve the lack into the trained model 222. If a question similar to the output result (a question to acquire the necessary information) is found in the user information DB, it selects this question. On the other hand, if there are no similar questions in the user information DB, the output result will be generated into a question that is easy for user U to understand based on the various information about user U described above.

Furthermore, when the information processing unit 246 analyzes the multiple requirement definitions output in the previous process and detects any inconsistencies between the multiple requirement definitions, the question providing unit 242 selects and generates additional questions for user U to resolve the inconsistencies. For example, assume that user U presents two requirements as instructions (answers): "The system must operate 24 hours a day, 365 days a year" and "I would like to perform regular system maintenance." As a result of analyzing these requirements, the information processing unit 246 detects that there is a contradiction between the system's operating time and maintenance time. In this case, in order to resolve the inconsistency, the question providing unit 242 generates and provides to user U, for example, a question such as, "What specific operations do you have in mind to achieve both 24 hours a day, 365 days a year system operation and regular maintenance? For example, please let us know if you have any requests or plans, such as the time periods when system downtime during maintenance is acceptable or the implementation of non-stop maintenance through system redundancy."

As another example, if user U provides an instruction (answer) that "the system must maintain a high level of security" and at the same time "I want to allow access without user authentication," the information processing unit 246 detects an inconsistency between the security requirements and the access requirements. In response, the question providing unit 242 generates a question such as, "What specific security measures are you considering to maintain a high level of security while allowing access without user authentication? For example, we could consider limiting the information that can be viewed, restricting the IP address of the access source, or anonymizing the data. Would you consider these?"

Furthermore, if user U provides instructions (answers) such as "The system is intended for use on mobile devices" and "It will need to handle a large number of high-resolution images," the information processing unit 246 detects an inconsistency between the characteristics of the mobile device and the handling of a large number of high-resolution images. In this case, the question providing unit 242 generates a question such as, "It seems that the system is intended for use on mobile devices, but will need to handle a large number of high-resolution images. Are you concerned about the amount of data communication or display speed in a mobile environment? For example, there are countermeasures such as image compression, use of thumbnail display, and loading images as needed, but please let us know your requests and expected behavior."

In this embodiment, as an example, when the information processing unit 246 performs analysis and detects the presence or absence of an inconsistency between multiple requirement definitions, the question providing unit 242 inputs the details of the inconsistency and a prompt that instructs the output of a question to acquire the information necessary to resolve the inconsistency into the trained model 222, and selects a question similar to the output result (a question to acquire the necessary information) if it is found in the user information DB. On the other hand, if there is no similar question in the user information DB, the question providing unit 242 generates the output result into a question that is easy for user U to understand based on the various information about user U described above.

Furthermore, when presenting a question to user U, the question providing unit 242 also has the function of providing background information related to the question (see background information BI in Figure 5 as an example of background information). Specifically, in the process of generating or selecting a question to provide to user U, the question providing unit 242 collects background information related to the question. This background information is obtained from the acquired information DB 224 (prompts and instructions previously entered by user U), the user information DB 226 (user U's account information and attributes, past response history), the processing result DB 230 (requirement definitions generated to date, their analysis results, and data related to missing information and inconsistencies), etc. The question providing unit 242 references this data and extracts background information related to the question using the trained model 222. First, it analyzes related past requirement definitions, response history, and analysis results to identify information useful to user U. Then, based on the extracted information, it generates background information in a format that is easy for user U to understand. At this time, the expression and level of detail of the background information are adjusted taking into consideration attributes of user U, such as their field of expertise, level of understanding, and job title, which are stored in user information DB 226. The background information generated in this way is integrated with the question and presented to user U. This enables user U to more deeply understand the intent and purpose of the question and provide an accurate answer. For example, if the question providing unit 242 generates a question such as "Please tell me in detail about the main functional requirements of the new system," the question providing unit 242 takes into consideration past answers and issues and provides the question to user U with background information such as "From your answers so far, it appears that the system you are currently using faces challenges in terms of efficient inventory management and real-time analysis of sales data. Additionally, the sales department has requested access via mobile devices. Taking these points into consideration, please tell us specifically about the functions you consider important in the new system and the areas you would like to improve."

The information acquisition unit 244 has the function of acquiring an answer from user U to a question provided by the question providing unit 242 (see answer A in Figure 5 as an example of an answer, including answers to additional questions), as well as acquiring various information required for processing by the question providing unit 242 and the information processing unit 246. Specifically, the information acquisition unit 244 performs the following processing. First, in acquiring an answer from user U, the information acquisition unit 244 receives the answer to the question provided to user U by the question providing unit 242 and stores the content of the answer in the processing result DB 230 of the storage unit 22. User U's answer is written in natural language, and the information acquisition unit 244 saves the answer in an appropriate format. The information acquisition unit 244 also acquires metadata such as user U's account information and the timestamp of the answer, and records this in association with the answer.

Next, in acquiring various pieces of information, the information acquisition unit 244 acquires data required for processing by the question provision unit 242 and the information processing unit 246. Specifically, it acquires information stored in the memory unit 22 or various data that can be acquired from outside using API integration, etc.

The information processing unit 246 has the function of generating a requirement definition using the trained model 222 based on the answers obtained from the user U. Specifically, the information processing unit 246 uses the answers from the user U obtained by the information acquisition unit 244 as input data and instructs the trained model 222 to generate a requirement definition. Then, it acquires information about the requirement definition generated by the trained model 222.

The information processing unit 246 also has the function of generating an evaluation and the basis for that evaluation based on predetermined evaluation criteria for the answer obtained from user U (see, for example, evaluation E and evaluation basis BE1 in Figure 6). Examples of these predetermined evaluation criteria include the specificity, validity, consistency, and comprehensiveness of the answer. First, to evaluate the specificity of the answer, it determines whether the answer contains specific numerical values, examples, and detailed explanations. For example, it can determine whether abstract expressions are overused and encourage the provision of specific information. Next, the validity evaluation determines whether the answer is technically and logically correct and realistic. This is done by comparing the answer with information from an existing technical knowledge base and industry standards to check for inconsistencies or unrealistic requirements. The consistency evaluation determines whether the answer is consistent with past answers and existing requirement definitions by comparing it with data accumulated in the processing result DB 230. The comprehensiveness evaluation determines whether the answer sufficiently contains the information necessary for the question and whether it includes expected answer items. The information processing unit 246 generates evaluation results and their basis based on these evaluation criteria and provides them to user U as feedback. Furthermore, the information processing unit 246 causes the question providing unit 242 to provide questions on whether or not to correct the answer entered during this feedback and whether or not to proceed to the next question (see questions CA and PQ in FIG. 6 as examples). This allows the user U to review the content of their answer, and make corrections or provide additional information as necessary, or to temporarily skip correcting the answer. Note that the evaluation method described above is an example and is not limited to this.

Furthermore, the information processing unit 246 has the function of analyzing the generated requirements definition and determining whether there is insufficient information in at least one of the questions and answers (see evaluation basis BE1 in Figure 6 as an example). If the analysis of the requirements definition identifies missing information or areas where details are lacking and determines that these need to be supplemented, the information processing unit 246 communicates this to the question providing unit 242 and executes processing to provide additional questions. The information processing unit 246 also has the function of determining whether there are inconsistencies between the multiple generated requirements definitions (see evaluation basis BE2 in Figure 7 as an example). If there are contradictions or discrepancies between the requirements definitions, it detects them and determines that an inconsistency exists. If an inconsistency is detected, the information processing unit 246 communicates this information to the question providing unit 242 and executes processing to provide additional questions to resolve the inconsistency.

The output unit 248 controls the output of the processing results of the question providing unit 242 and the information processing unit 246 to the user terminal 14. That is, the output unit 248 outputs at least one of the question, additional question, and background information that are the processing results of the question providing unit 242 to the user terminal 14. The output unit 248 also outputs at least one of the request definition that is the processing result of the information processing unit 246, an evaluation of the answer based on predetermined evaluation criteria, and the basis for the evaluation to the user terminal 14.

(Processing Executed by Information Processing System 10)
Next, the operation of the information processing system 10 will be described. Fig. 4 is a flowchart showing an example of the flow of processing by the information processing system 10. The processor 120 reads out the program 220 stored in the storage 124, expands it in the memory 122, and executes it, thereby performing processing. Although not shown, when the processor 120 receives operation information to terminate the operation of the information processing system 10 or operation termination information from the user terminal 14 during the ongoing determination processing (these will be simply referred to as "termination operations"), the processor 120 terminates processing based on the program 220 that is currently being processed.

The processor 120 acquires a prompt to perform a request definition input from the user U, etc. (step S100), and then acquires and/or generates a question according to the content of the prompt, and outputs the question to the user terminal 14, thereby providing the question to the user U (step S102). The processor 120 then outputs background information for the question to the user U to the user terminal 14, thereby providing the information to the user U (step S104), and also displays an answer input screen for the provided question on the user terminal 14 (step S106).

The processor 120 determines whether or not an answer to the provided question has been obtained from the user U (step S108). If an answer cannot be obtained (step S108: NO), the processor 120 proceeds to step S106. On the other hand, if an answer has been obtained (step S108: YES), the processor 120 provides the user U with an evaluation of the answer based on predetermined criteria and the basis for the evaluation by displaying them on the user terminal 14 (step S110).

The processor 120 prompts the user U as to whether or not to revise the answer by displaying the answer on the user terminal 14, and determines whether or not the user U has input an instruction to revise the answer (step S112). If an instruction to revise has been received (step S112: YES), the processor 120 proceeds to step S106. On the other hand, if an instruction to revise has not been received (step S112: NO), the processor 120 generates a requirement definition based on the input answer (step S114) and performs various analyses of the question, the answer to that question, and the requirement definition (step S116). Note that in this embodiment, this analysis involves determining whether or not there is a lack of information in at least one of the question and answer, and determining whether or not there is an inconsistency between the multiple requirement definitions that have been generated, but is not limited to these.

The processor 120 determines whether or not there is insufficient information as a result of the analysis (step S118). If there is insufficient information (step S118: YES), the processor 120 proceeds to step S102 to provide additional questions. On the other hand, if there is no insufficient information (step S118: NO), the processor 120 determines whether or not there are inconsistencies between the requirement definitions if there are multiple requirement definitions generated as a result of the analysis (step S120). If there are inconsistencies (step S120: YES), the processor 120 proceeds to step S102 to provide additional questions to resolve the inconsistencies. On the other hand, if there are no inconsistencies (step S120: NO), the processor 120 outputs information about the generated requirement definition to the user terminal 14 for display (step S122; as an example, see requirement definition RD, which is information about the requirement definition in Figure 10).

The processor 120 determines whether an instruction to end the process has been issued by the user U (step S124). If an instruction to end the process has not been issued (step S124: NO), the processor 120 proceeds to step S102. On the other hand, if an instruction to end the process has been issued (step S124: YES), the processor 120 ends the process based on the program 220. Note that the above-mentioned steps S102 and S104 correspond to the "question providing step" set forth in claim 1, steps S106 and S108 correspond to the "answer obtaining step" set forth in claim 1, steps S110 and S112 correspond to the "output generating step" set forth in claims 1 and 3, and steps S116 and S118 correspond to the "verification step" set forth in claim 4.

(Operation and effect of the first embodiment)
The information processing system 10 according to this embodiment executes a question provision step for selecting and/or generating questions to obtain requirement definitions; an answer acquisition step for acquiring answers to the questions; and an output generation step for generating requirement definitions from the answers using a trained model 222 that learns at least text data and generates requirement definitions inferred from the answers when information including the answers is input. In other words, the trained model 222 infers and extracts potential requirement definitions from the answers of the user U, thereby enabling the collection of requirements not explicitly stated by the user U and overlooked needs. This is because the trained model 222 has learned a large amount of text data and is capable of deeply understanding the content of the user U's answers using natural language processing technology and inferring the underlying intentions and requests. As a result, information obtained from the user U is comprehensively collected, including not only the presented answers but also the background and related information. This makes it possible to identify requirement definitions with no limited scope of application, thereby improving the accuracy and comprehensiveness of requirement definitions in system development. This advanced requirements definition collection process is expected to reduce missing requirements and unclear points in the early stages of development, contributing to the efficiency and quality improvement of the entire system development. Furthermore, by discovering the latent needs of users U early on, it is possible to reduce the risk of subsequent specification changes and additional development, and also to increase the success rate of the project. This will support system development by identifying requirements definitions that are not limited in scope.

In addition, the information processing system 10 includes a process for providing background information about the question along with the question in the question generation step, allowing user U to gain a deeper understanding of the intent, purpose, and related context of the question, making it easier to provide a more specific and detailed answer that is tailored to their own situation and needs. From another perspective, providing background information also allows user U to become aware of potential requirement definitions that they themselves were unaware of, or needs that they may have overlooked. For example, by providing information such as current business challenges, market trends, and constraints of existing systems as the background to a question about a specific system function, user U can consider requirement definitions from a broader perspective. This makes it possible to effectively identify requirement definitions that are not limited in scope, thereby effectively supporting system development.

Furthermore, since the information processing system 10 includes a process for generating at least one of an evaluation of the answer based on predetermined evaluation criteria and the basis for that evaluation in the output generation step, user U can understand how his or her answer was evaluated and the basis for that evaluation. By receiving feedback on the evaluation results and the basis, user U recognizes shortcomings in his or her answer and areas for improvement, and is encouraged to provide more specific and accurate information. This improves the quality of the information provided by user U and increases the accuracy and comprehensiveness of the requirements definition. As a result, it becomes possible to more effectively identify requirements definitions that are not limited in scope, and system development can be more effectively supported. In particular, by evaluating the answers and providing the basis for those evaluations, communication with user U is deepened, making it possible to clarify latent needs and overlooked requirements. This is expected to reduce missing requirements and misunderstandings in the early stages of development, and increase the success rate of system development projects.

Furthermore, the information processing system 10 includes a verification step of analyzing the requirements definition generated by the output generation step and detecting missing information in at least one of the questions and answers, and, if missing information is detected in the verification step, re-executing the question generation step to provide additional questions to acquire the missing information and re-executing the answer acquisition step to acquire the additional answers. In this way, the information processing system 10 verifies the generated requirements definition and repeats the process of additional questions and answers to supplement the missing information. This action further enriches the information obtained from user U, improving the comprehensiveness and accuracy of the requirements definition. This enables more effective identification of requirements definitions that are not limited in scope, and more effective support for system development. In particular, detecting missing information and asking additional questions to supplement it clarifies user U's latent needs and overlooked requirements, resulting in a reduction in missing requirements and unclear points in the early stages of development.

Furthermore, the information processing system 10 determines whether there are any inconsistencies between the multiple requirements definitions generated in the output generation step. If an inconsistency is determined, the system re-executes the question generation step to provide additional questions to acquire information necessary to resolve the inconsistency, and re-executes the answer acquisition step to acquire additional answers. This allows for the effective collection of information needed to resolve the inconsistency between the requirements definitions. Specifically, if a contradiction or discrepancy is detected between the generated requirements definitions, the information processing system 10 automatically identifies the inconsistency and provides appropriate additional questions to the user U. This allows the user U to reconfirm or provide a detailed explanation of the information causing the inconsistency. The information processing system 10 then updates the requirements definitions using the newly acquired answers to resolve the inconsistency. This action improves the consistency and accuracy of the requirements definitions obtained from the user U, enabling more effective identification of requirements definitions with unlimited scope. As a result, system development can be more effectively supported. In particular, automating the process of early detection and resolution of inconsistencies in requirements definitions can be expected to reduce missing requirements and misunderstandings in the early stages of development, thereby increasing the project success rate.

Second Embodiment
Next, an information processing system 40 according to a second embodiment of the present invention will be described with reference to Figures 8 to 11. The information processing system 40 according to the second embodiment has the same basic configuration as the first embodiment, and is characterized in that it determines the dependencies between multiple generated requirement definitions, acquires and outputs information about the dependencies, and provides questions including confirmation of whether previously generated requirement definitions have been updated. Note that the same components as those in the first embodiment are designated by the same reference numerals, and their description will be omitted.

(Functional configuration)
The functional configuration of the server 50 as an information processing device in the information processing system 40 will be described. FIG. 8 is a diagram showing an example of the functional configuration of the server 50. When executing various programs, the server 50 realizes various functions using the same hardware resources as the server 12 of the first embodiment. The server 50 has a communication unit 20, a storage unit 60, and a control unit 70 as the functional configuration realized by the server 50. Each functional configuration is realized when the processor 120 reads and executes a program 600 stored in the memory 122 or the storage 124.

The memory unit 60 is realized by the memory 122 and storage 124. The memory unit 60 stores a program 600, a trained model 222, an acquired information DB 224, a user information DB 226, a user information DB, and a processing result DB 230.

The control unit 70 is realized by the processor 120 reading and executing the program 600 from the memory 122 (see Figure 2) and working in cooperation with other hardware components. The control unit 70 includes a question providing unit 700, an information acquisition unit 702 as an answer acquisition unit, an information processing unit 704 as an output generation unit, and an output unit 706.

Similar to the question providing unit 242 of the first embodiment, the question providing unit 700 has the function of providing questions to the user U, and performs at least one of selecting and generating questions. Furthermore, the question providing unit 700 has the function of providing questions that include confirmation of whether previously generated requirement definitions have been updated. Specifically, the question providing unit 700 generates or selects questions that, as necessary, include confirmation of whether these requirement definitions are still valid or require changes or updates, and provides these to the user U, based on the information processing unit 704's reference to the past requirement definitions stored in the memory unit 22 and the processing result DB 230.

Like the information acquisition unit 244 in the first embodiment, the information acquisition unit 702 has the function of acquiring answers (including answers to additional questions) from the user U to the questions provided by the question provision unit 242, as well as acquiring various information required for processing by the question provision unit 242 and the information processing unit 246. Furthermore, the information acquisition unit 702 has the function of performing processing to acquire a deletion instruction for a requirement definition. Specifically, the information acquisition unit 702 accepts an instruction from the user U to delete a specific requirement definition, and records the deletion instruction in the processing result DB 230 of the storage unit 22. This allows the user U to appropriately delete requirement definitions that are no longer necessary, thereby maintaining the currency and consistency of the requirement definitions. Based on the acquired deletion instruction, the information acquisition unit 702 works in cooperation with the information processing unit 246 to execute deletion processing for the corresponding requirement definition.

The information processing unit 704, like the information processing unit 246 in the first embodiment, executes processing based on the prompt. The information processing unit 704 also stores the processing results corresponding to the acquired information DB 224 in the storage unit 60, linking the acquired information DB 224 with the user information DB 226.

Like the information processing unit in the first embodiment, the information processing unit 704 has the function of generating requirement definitions using the trained model 222 based on answers obtained from user U. It also has the function of generating an evaluation of the answers obtained from user U based on predetermined evaluation criteria and the basis for that evaluation, the function of analyzing the generated requirement definitions to detect information deficiencies, and the function of determining whether there are inconsistencies between multiple requirement definitions. The information processing unit 704 also has the function of determining dependencies between the multiple generated requirement definitions, and, if there are dependencies, acquiring and outputting information about the dependencies. Specifically, the information processing unit 704 analyzes the associations and dependencies between the generated requirement definitions, and stores these relationships in the storage unit 60 in the form of text data, graphs (not shown), or the like, and presents them to user U. This makes it possible to clarify the priority and scope of impact of requirement definitions and support project planning and risk management.

The information processing unit 704 also performs a process to set a specific requirement definition from among multiple requirement definitions as a reference requirement definition, and then performs a reference determination process to determine at least one of the following: information deficiencies, inconsistencies, and updates for other requirement definitions based on the reference requirement definition. Specifically, as shown in FIG. 10 , the user terminal 14 displays requirement definitions RD (including previously generated ones) and displays lock/unlock instruction buttons LB that can be operated for each specified category (e.g., chapters in this embodiment), and accepts instructions from the user U regarding which requirement definitions to lock or unlock. A requirement definition that is instructed to be locked becomes the reference requirement definition. Locked requirement definitions remain unchanged, and only related requirement definitions are subject to change or update. On the other hand, if an instruction to unlock a locked requirement definition is issued, the requirement definition is excluded from the reference requirement definition, making it clear that it is subject to change or update. In other words, unlocking is performed on the locked requirement definition. The information processing unit 704 re-evaluates the unlocked requirement definition for information deficiencies, inconsistencies, and the need for updates with other requirement definitions. The information processing unit 704 then notifies the question providing unit 700 as necessary, and generates and provides additional questions, thereby collecting information to review the contents of the unlocked requirements definition and regenerating (updating) it (see FIG. 11). In this way, unlocking is performed on locked requirements definitions, and the unlocked requirements definitions become the subject of re-examination and updating. Based on locking and unlocking instructions from user U, the information processing unit 704 flexibly manages requirements definition standards and changes, thereby improving the accuracy, consistency, and flexibility of requirements definitions in system development. The above-mentioned standard requirements definitions correspond to the "standard requirements" set forth in claim 8.

Furthermore, the information processing unit 704 performs processing to acquire a deletion instruction for the requirement definition, and upon acquiring the deletion instruction, executes processing to delete the requirement definition. Specifically, it receives a deletion instruction from user U via the information acquisition unit 702, and deletes the corresponding requirement definition from the processing result DB 230, etc. Furthermore, if a dependency or inconsistency arises as a result of the deletion, it re-determines the dependency and detects the inconsistency, and updates the related requirement definition as necessary. In this way, the consistency and up-to-dateness of requirement definitions can be maintained, making it possible to smoothly proceed with the system development process.

Furthermore, the information processing unit 704 has the function of determining whether to acquire requirement definitions from the storage unit at a predetermined timing. Predetermined timing includes when it is determined that requirement definitions need to be reacquired based on the content of a prompt entered by the user, or when a scheduled timing is set for periodic review. As a specific example, if user U enters "I would like to review the functional requirements in light of recent market trends" in a new prompt, the information processing unit 704 analyzes the content and determines that existing requirement definitions need to be acquired from the storage unit. It is also possible to automatically reacquire requirement definitions when the project transitions between phases or at regular intervals to check their latest status. The information processing unit 704 then notifies the question providing unit 700 to provide questions about the acquired requirement definitions, including whether they have been updated, as described above.

The output unit 706, like the output unit 248 in the first embodiment, controls the output of the processing results of the question providing unit 242 and the information processing unit 246 to the user terminal 14. Furthermore, when dependencies between multiple generated requirement definitions are recognized and information regarding the dependencies is generated, the output unit 706 has the function of outputting this information to the user terminal 14. Specifically, it obtains information regarding the dependencies between requirement definitions analyzed by the information processing unit 704 and displays it as text data or graph data so that the user can confirm it.

(Processing Executed by Information Processing System 40)
Next, the operation of the information processing system 40 will be described. Fig. 9 is a flowchart showing an example of the flow of processing by the information processing system 40. The processor 120 reads out the program 600 stored in the storage 124, expands it in the memory 122, and executes it, thereby performing processing. Note that the same processes as those in the first embodiment are assigned the same reference numerals, and descriptions thereof will be omitted.

After processing step S100, processor 120 analyzes the contents of the prompt (step S200) and determines whether to acquire the request definition stored in memory unit 60 at a predetermined timing (step S202). If not (step S202: NO), processor 120 proceeds to step S102. On the other hand, if acquisition is required (step S202: YES), processor 120 acquires the request definition determined to be acquired from memory unit 60 (step S204), and presents user U with a question including confirmation as to whether the acquired request definition is still valid or requires modification or updating (step S206), and proceeds to processing step S104.

After processing step S114, the processor 120 stores the generated request definition in the storage unit 60 (processing result DB 230) (step S208), and determines whether an instruction has been received from user U as to which request definition to lock or unlock (step S210). If an instruction has not been received (step S210: NO), the processor 120 proceeds to step S214, which will be described later. On the other hand, if an instruction has been received (step S210: YES), the processor 120 reflects the locking or unlocking of the requested request definition based on the instruction, and then regenerates the request definitions other than the one reflected in the locking, using the same processing as step S114 (step S212).

The processor 120 determines whether an instruction to delete a specific request definition has been received from the user U (step S214). If an instruction has not been received (step S212: NO), the processor 120 proceeds to step S218, which will be described later. On the other hand, if an instruction has been received (step S212: YES), the processor 120, based on the instruction, deletes the request definition that is the target of the instruction from the processing result DB 230, etc., and reflects the instruction (step S216).

The processor 120 analyzes the question, the answer to the question, and/or the requirement definition (step S218). Note that the analysis in this embodiment not only includes the analysis in the first embodiment, but also includes determining the dependencies between the multiple requirement definitions that have been generated. The analysis results are stored in the memory unit 60, and then processing proceeds to step S118.

If no inconsistency is found in step S120 (step S120: NO), the processor 120 outputs information about the generated requirement definitions to the user terminal 14 for display (step S220). Note that, in step S220, if it is determined in the aforementioned step S218 that there are dependencies between multiple requirement definitions, information about the dependencies is also output. The processor 120 then proceeds to step S124. Note that steps S210, S212, S218, S118, and S120 correspond to the "criteria determination process" set forth in claim 8.

(Operation and effect of the second embodiment)
The information processing system 40 according to this embodiment is similar in configuration to the first embodiment, except that it determines dependencies between multiple generated requirement definitions, acquires and outputs information about the dependencies, and provides questions including confirmation of whether previously generated requirement definitions have been updated. Therefore, it achieves the same effects as the first embodiment. Furthermore, because the information processing system 40 determines dependencies between multiple generated requirement definitions and acquires and outputs information about the dependencies, the user U can clearly understand the relationships and priorities between requirement definitions. Specifically, the user U can understand how a requirement definition affects other requirement definitions or whether a specific requirement definition depends on other requirement definitions. This leads to improved quality of responses from the user U, enabling the identification of requirement definitions with unlimited applicability, and further supporting system development.

Furthermore, the information processing system 40 stores requirements definitions in the storage unit 60 and retrieves them from the storage unit 60 at predetermined times to provide questions, including confirmation of whether the requirements definitions have been updated. This allows the system to periodically check whether existing requirements definitions are up to date and appropriately prompt the user U regarding the need for updates. Predetermined timing includes user instructions or the passage of a certain period of time. For example, when a project phase progresses or market trends change, the system automatically retrieves previous requirements definitions from the storage unit 60 and prompts the user with a question such as, "Are there any changes or additions to the requirements definitions you previously provided?" This allows the user to communicate new needs and changes to the system, facilitating the updating of requirements definitions. As a result, it becomes possible to identify requirements definitions with unlimited scope, further supporting system development. Specifically, the system's ability to respond quickly to user requests and market changes contributes to more efficient development processes and improved product quality. Furthermore, preventing missing requirements definitions and development based on outdated information reduces overall project risk and increases the success rate.

Furthermore, the information processing system 40 includes a process for setting a specific requirement definition from among multiple requirement definitions as a reference requirement definition (locking process) and a criteria determination process for determining at least one of information deficiencies, inconsistencies, and updates for other requirement definitions based on the reference requirement definition. Therefore, user U can set a specific requirement definition from among multiple requirement definitions as the reference requirement definition and perform locking on it. The locking process prevents the reference requirement definition from being changed. This function allows user U to clearly reflect their intention, such as "I do not want this requirement definition to be changed," in the system. Furthermore, the information processing system 40 sets the reference requirement definition locked by user U as a reference point. This ensures that the criteria are clear and consistent when comparing and verifying other requirement definitions with the reference requirement definition. Specifically, the criteria determination process is performed based on the reference requirement definition to determine at least one of information deficiencies, inconsistencies, and the need for updates for other requirement definitions. As a result, requirement definitions that accurately reflect user U's intentions and have no limited scope of application can be identified. This leads to more effective support for system development.

In addition, the information processing system 40 includes a process for providing questions to resolve at least one of the following cases based on the results of the criteria determination process: when information in other requirements definitions is insufficient, there is an inconsistency, or an update is necessary, the information processing system 40 can collect the necessary information from the user U and resolve or update any missing or inconsistent requirements definitions or outdated information. This action makes it possible to timely supplement or resolve missing or inconsistent information in requirements definitions, as well as the need for updates. This makes it possible to identify requirements definitions that are not limited in scope, further supporting system development. In particular, by including a process for verifying other requirements definitions based on the benchmark requirements definition and proactively resolving problems, the accuracy and comprehensiveness of requirements definitions is improved, contributing to the efficiency and quality improvement of the development process.

In the second embodiment described above, the information processing system 40 is configured to determine the dependencies between multiple generated requirement definitions, acquire information about the dependencies, and output it for presentation to the user U. However, this is not limiting; the information about the dependencies may be acquired and used for internal processing without being output for presentation to the user U. One example of this internal processing is a process in which, when a specified requirement definition is changed, information about the dependencies is referenced and other requirement definitions that are dependent on it are automatically changed. This makes it possible to implement changes to requirement definitions without any omissions.

(Third embodiment)
Next, an information processing system 80 according to a third embodiment of the present invention will be described using Figures 12 to 14. The information processing system 80 according to the third embodiment has the same basic configuration as the first and second embodiments, and is characterized in that it includes a process for generating documents in which requirement definitions are classified according to predetermined criteria, and a process for inputting the requirement definitions for each document classification into the trained model 222 and outputting requirement definitions as secondary requirements. Note that components identical to those in the first and second embodiments are assigned the same reference numerals and their description will be omitted. Furthermore, the requirement definition in this embodiment differs from the requirement definition in the previously described embodiments in that it specifically summarizes what functions a system developed based on the requirement definition should have and how it should be.

(Functional configuration)
The functional configuration of the server 82 as an information processing device in the information processing system 80 will be described. Fig. 12 is a diagram showing an example of the functional configuration of the server 82. When executing various programs, the server 82 realizes various functions using the same hardware resources as the server 12 of the first embodiment. The server 82 has a communication unit 20, a storage unit 90, and a control unit 1000 as the functional configuration realized by the server 82. Each functional configuration is realized when the processor 120 reads and executes a program 900 stored in the memory 122 or the storage 124.

The memory unit 90 is realized by the memory 122 and storage 124. The memory unit 90 stores a program 900, a trained model 222, an acquired information DB 224, a user information DB 226, a user information DB, and a processing result DB 230.

The control unit 1000 is realized by the processor 120 reading and executing the program 900 from the memory 122 (see Figure 2) and working in cooperation with other hardware components. The control unit 1000 includes a question providing unit 700, an information acquisition unit 1002 as an answer acquisition unit, an information processing unit 1004 as an output generation unit, and an output unit 1006.

Like the information acquisition unit of the first embodiment, the information acquisition unit 1002 has the function of acquiring answers (including answers to additional questions) from the user U to the questions provided by the question provision unit 242, as well as acquiring various information required for processing by the question provision unit 242 and the information processing unit 1004. Furthermore, the information acquisition unit 1002 acquires a document in which the requirement definitions generated by the information processing unit 1004 are classified according to predetermined criteria, i.e., a requirement definition document (details will be described later).

The information processing unit 1004 has the same functions as the information processing unit 246 in the first embodiment and the information processing unit 704 in the second embodiment. Furthermore, the information processing unit 1004 has the function of generating a requirements definition document (see the requirements definition RD in Figure 10 for an example) as an example of a document in which requirements definitions are classified according to predetermined criteria. Specifically, as an example, assume that requirement definitions for a "user authentication function" and a "product search function" exist. If a new requirement definition for a "recommendation function" is added to these, the information processing unit 1004 integrates these requirement definitions to create a consistent requirements definition document. First, the existing requirement definitions are carefully examined to check for any inconsistencies or overlaps with the new requirement definition. Next, each requirement definition is classified according to predetermined criteria and organized into chapters such as "Chapter 1: System Overview," "Chapter 2: Functional Requirements," and "Chapter 3: Non-Functional Requirements." In this embodiment, these predetermined criteria are used to classify the requirements into chapters by content. Specifically, "Chapter 1: System Overview" describes the system's purpose, background, and overall concept. "Chapter 2: Functional Requirements" details the requirements for specific functions the system should have, such as user authentication, product search, and cart functionality. "Chapter 3: Non-Functional Requirements" describes requirements related to non-functional qualities and constraints, such as performance, security, scalability, and reliability. Note that these are not the only prescribed criteria. Another specific criterion is "classification by priority." Requirements are classified into "high priority," "medium priority," and "low priority" based on their importance and urgency. For example, high priority includes requirements that must be implemented as basic system functionality. Medium priority includes requirements that are important but can be addressed in the next phase, while low priority includes requirements that should be considered for future expansion or as options. Furthermore, "classification by function" can also be used as a specific criterion. For example, "user management functionality" includes requirements for user registration, login, and password reset. "Product management functionality" summarizes requirements related to product registration and editing and inventory management. "Order processing functionality" includes adding items to the cart, completing the purchase process, and displaying order history. This makes it easier to organize requirement definitions by function and divide them up among the responsible departments or teams. Another specific criterion is "classification by user." By dividing requirement definitions into those for general users and those for administrators, it is possible to clarify the functions that meet the needs of each user group. For general users, requirement definitions related to browsing the site, searching for products, and purchasing procedures are included, while for administrators, requirement definitions related to managing the entire site, managing user accounts, updating product data, etc. These specified criteria may be a combination of the other specific criteria listed above, or may be other criteria.

The information processing unit 1004 is also capable of generating detailed content for each requirements definition by utilizing the trained model 222. Specifically, the requirements definition and prompts that instruct the trained model 222 to supplement it with specialized knowledge and appropriate expressions are input, and an output is obtained. During this process, the prompts are adjusted to reflect the intentions of user U and the overall objectives of the system. The content of each generated chapter is integrated by the information processing unit 1004 and compiled into a requirements definition document. Finally, the created requirements definition document is provided to user U via the output unit 1006 and stored in the memory unit 60 (processing result DB 230).

The information processing unit 1004 also analyzes the generated requirements definition document, inputs the requirements definitions of the requirements definition document into the trained model 222 for each classification of the requirements definition document, and performs processing to output requirements definitions, which are secondary requirements. This enables efficient and high-quality requirements definitions to be generated from the requirements definitions. That is, the information processing unit 1004 first performs processing for each classification of the requirements definition document RD. Specifically, as shown in FIG. 15, the requirements definitions are processed by chapter, such as "Chapter 1: System Overview (corresponding to node "1" in the figure)," "Chapter 2: Functional Requirements (corresponding to node "2" in the figure)," and "Chapter 3: Non-Functional Requirements (corresponding to node "3" in the figure)." First, the requirements definition of "Chapter 1: System Overview" is input into the trained model 222, and a requirements definition for the entire system is generated (this corresponds to the edge from RD to node 1 in the figure). At this time, the information processing unit 1004 sets and inputs appropriate prompts so that a requirements definition can be obtained from the requirements definition RD, and adjusts the trained model 222 to obtain the expected output. In this embodiment, the prompts are output to the user terminal 14 to be presented to the user U, and at least one of changes, modifications, and additions to the prompts by the user U can be accepted and acquired. However, this is not limited to this, and the configuration may be such that the processing is performed automatically without presenting the prompts to the user U. Next, when processing "Chapter 2: Functional Requirements," the requirements definitions for "Chapter 2: Functional Requirements" are input into the trained model 222 along with appropriate prompts as described above to generate a requirements definition for the functional requirements (the edge from RD to node 2 in the figure corresponds to this). At the same time, an edge (dependency relationship) is constructed to reference the requirements definition for "Chapter 1: System Overview" that was previously generated (the edge from node 1 to node 2 in the figure corresponds to this). As a result, the information in "Chapter 1" is taken into account when generating "Chapter 2: Functional Requirements," and a consistent requirements definition is generated. Similarly, when processing "Chapter 3: Non-functional Requirements," the requirements definition for "Chapter 3: Non-functional Requirements" is entered into the trained model 222 along with the appropriate prompts as described above to generate a requirements definition for functional requirements (the edge from RD to node 3 in the diagram corresponds to this), while at the same time, edges are constructed between the requirements definitions for "Chapter 1" and "Chapter 2" previously generated (the edge from node 1 to node 3 and the edge from node 2 to node 3 in the diagram correspond to this). This allows the aforementioned information to be utilized when generating "Chapter 3: Non-functional Requirements," resulting in a consistent requirements definition for the entire system. This sequential processing and edge construction links information between each chapter, making it possible to efficiently generate high-quality requirements definition documents that are consistent with each other.

Similar to the output unit 248 of the first embodiment and the output unit 706 of the second embodiment, the output unit 1006 controls the output of the processing results of the question providing unit 242 and the information processing unit 246, as well as information regarding the dependencies between multiple requirement definitions, to the user terminal 14. Furthermore, the output unit 706 can be controlled to output the requirement definition document and the requirements definition document to the user terminal 14.

(Processing Executed by Information Processing System 80)
Next, the operation of the information processing system 80 will be described. Fig. 13 is a flowchart showing an example of the flow of processing by the information processing system 80. The processor 120 reads out the program 600 stored in the storage 124, expands it in the memory 122, and executes it, thereby performing processing. Note that the same processes as those in the first and second embodiments are denoted by the same reference numerals, and descriptions thereof will be omitted.

After processing step S200, the processor 120 determines whether an instruction to output the requirements definition document has been received from the user U (step S300). If an output instruction has not been received (step S300: NO), the processor 120 proceeds to step S124. On the other hand, if an output instruction has been received (step S300: YES), the processor 120 outputs the requirements definition document and stores it in the memory unit 90 (step S302).

The processor 120 retrieves the requirements definition document stored in the memory unit 90 (step S304), analyzes the requirements definition document (step S306), outputs a prompt to the user terminal 14 for defining requirements from the requirements definition document set by the processor 120, and executes processing to display a screen on the user terminal 14 for accepting at least one of changes, modifications, and additions to the prompt (step S308).

The processor 120 determines whether or not to receive at least one instruction from the user U to change, modify, or add to the prompt (step S310). If no instruction has been received (step S310: NO), the processor 120 proceeds to step S314, which will be described later. On the other hand, if an instruction has been received (step S310: YES), the processor 120 reflects the content of the instruction in the prompt (step S312) and then generates a requirements definition for each category in the requirements definition document. As described above, the requirements definition is generated by sequential processing for each category (for example, for each chapter structure) and processing that involves edge construction as necessary.

The processor 120 determines whether the generation process of requirements definitions for all categories in the requirements definition document has been completed (step S316). If the process has not been completed (step S316: NO), the processor 120 proceeds to step S314. On the other hand, if the process has been completed (step S316: YES), the processor 120 executes a process to output the generated requirements definitions to the user terminal 14 for presentation to the user U (step S318).

The processor 120 stores the generated requirements definition in the storage unit 60 (processing result DB 230) (step S318), and determines whether an instruction has been received from the user U as to which requirements definition to lock or unlock (step S320). If no instruction has been received (step S320: NO), the processor 120 proceeds to step S324, described below. On the other hand, if an instruction has been received (step S320: YES), the processor 120 reflects the locking or unlocking of the requirements definition that is the target of the instruction based on the instruction (step S322).

The processor 120 determines whether an instruction to delete a specific requirements definition has been received from the user U (step S324). If an instruction has not been received (step S324: NO), the processor 120 proceeds to step S328, which will be described later. On the other hand, if an instruction has been received (step S324: YES), the processor 120, based on the instruction, deletes the requirements definition that is the target of the instruction from the processing result DB 230, etc., and reflects the instruction (step S326).

The processor 120 analyzes the question, the response to the question, and/or the requirement definition (step S328). Note that in this embodiment, analysis also includes determining the dependencies between the multiple requirement definitions that have been generated. Processing then proceeds to step S330.

As a result of the analysis, the processor 120 determines whether or not there is insufficient information (step S330). If there is insufficient information (step S330: YES), the processor 120 proceeds to step S314 to generate a new requirements definition. On the other hand, if there is no insufficient information (step S330: NO), the processor 120 determines whether or not there are inconsistencies between the requirements definitions if there are multiple requirements definitions generated as a result of the analysis (step S332). If there are inconsistencies (step S332: YES), the processor 120 proceeds to step S314 to generate a new requirements definition to resolve the inconsistencies. On the other hand, if there are no inconsistencies (step S332: NO), the processor 120 outputs information about the generated requirements definition to the user terminal 14 for display (step S334). Note that this output may not only include the requirements definition itself, but may also include the requirements definition organized in a prescribed format and output as a requirements definition document. Specifically, the processor 120 first retrieves predefined templates and formats (e.g., requirements definition documents, design specifications, project plans, etc.) from the storage unit 90 or external storage. Next, the processor 120 classifies the requirements definitions based on predetermined classification criteria (e.g., by function, priority, module, etc.). Based on the classification results, each requirements definition is assigned to a corresponding chapter or section of the document. The processor 120 then places the detailed content of each requirements definition in the appropriate location within the template. In this process, the processor 120 automatically generates headings and item numbers as necessary to organize the overall document structure. The processor 120 also applies predetermined formatting, such as font, paragraph, indentation, and bullet points, to the placed content. This improves the readability and consistency of the document. Finally, through the above process, a document conforming to a predetermined format and including the requirements definition is generated. This document is output as an electronic file (e.g., PDF, Word, HTML, etc.), stored in the storage unit 90, and made available to the user U via the output unit 1006. The "document" mentioned above includes information recorded on paper, electronic, or other recording media. After the above processing, processor 120 proceeds to step S124.

(Operation and effect of the third embodiment)
The information processing system 80 according to this embodiment has the same configuration as the first and second embodiments, except that it includes a process for generating a document in which requirement definitions are classified according to predetermined criteria and a process for inputting the requirement definitions into the trained model 222 for each document classification and outputting the requirement definitions. Therefore, the same operational effects as those of the first and second embodiments can be achieved. Furthermore, by including a process for generating a requirements definition document, which is a document in which requirement definitions are classified according to predetermined criteria, and a process for inputting the requirement definitions into the trained model 222 for each document classification and outputting the requirement definitions, detailed requirements definitions suited to each classification can be generated. In other words, if the entire requirements definition document is processed in one go rather than by document classification, the characteristics of each requirement definition are not fully reflected, and the content tends to be thin. This is because the trained model 222 may have a limit to the length or amount of information it can process at one time. If the entire requirements definition document is input in one go, the trained model 222 attempts to process the entire input evenly, distributing processing resources for each requirement definition. As a result, the trained model 222 may not fully understand the details and characteristics of each requirement definition, resulting in a superficial and shallow requirement definition. Furthermore, processing a large amount of information at once can make it difficult to evaluate the importance and relevance of information within the trained model 222. Important requirement definitions and special conditions are likely to be buried under other information and not be properly reflected. As such, the distribution of computational resources in the trained model 222 reduces overall quality. Furthermore, processing the entire requirements definition document in one go makes it difficult to accurately grasp the dependencies and relevance between each requirement definition. If the model cannot properly understand the interactions between each requirement definition, inconsistent requirements definitions may be generated. In contrast, by performing sequential processing by classification (chapter unit) as in this embodiment, the trained model 222 can concentrate computational resources on the requirements definitions in each chapter. This allows for the generation of requirements definitions that fully reflect the characteristics and details of each requirement definition. Furthermore, by referencing information generated in the previous chapter through edge construction, it is possible to maintain the consistency and integrity of requirements definitions between chapters. This allows for appropriate requirements definitions and requirement definition identification without limiting the scope of application, thereby further supporting system development.

In addition, the information processing system 80 also includes a process for converting requirements definitions into requirements definition documents organized in a specified format, thereby streamlining the management, sharing, and communication of requirements definitions. This enables a unified understanding and smoother communication among related stakeholders in a system development project.

In the first to third embodiments described above, the requirements are defined as requirements for system development, but they are not limited to this. They may also be other requirements, such as safety and design requirements in various projects, quality and cost requirements in product development, customer satisfaction and response speed requirements in the service industry, performance conditions and scope of responsibility requirements in legal contracts, learning outcomes and curriculum requirements in education, etc. In the third embodiment described above, secondary requirements are defined as requirements, but they are not limited to this. They may also be other requirements, such as secondary requirements for earthquake resistance and fire prevention performance derived from the safety requirement in various projects, secondary requirements for durability and ease of use derived from the quality requirement in product development, secondary requirements for staff response skills and service provision time derived from the customer satisfaction requirement in the service industry, and secondary requirements for the quality of teaching materials and individualized instruction derived from the learning outcome requirement in education.

<Additional Notes>
The present embodiment includes the following disclosure.

(Appendix 1)
An information processing method executed by an information processing device,
a question providing step of performing at least one of selecting a question for obtaining requirements and generating the question;
an answer acquisition step of acquiring an answer to the question;
an output generation step of generating the requirements from the answers using a trained model that learns at least text data and generates requirements estimated from the answers when information including the answers is input.

(Appendix 2)
the step of providing a question includes a process of providing background information on the question together with the question;
2. The information processing method according to claim 1.

(Appendix 3)
the output generation step includes a process of generating at least one of an evaluation of the answer based on predetermined evaluation criteria and a reason for the evaluation;
10. The information processing method according to claim 1 or 2.

(Appendix 4)
a verification step of analyzing the requirements generated by the output generation step and detecting missing information in at least one of the questions and the answers;
When a lack of information is detected in the verification step, the question providing step is executed again to provide an additional question for acquiring the lacking information, and the answer obtaining step is executed again to obtain the additional answer.
4. The information processing method according to any one of claims 1 to 3.

(Appendix 5)
determining whether or not there is an inconsistency among the plurality of requirements generated in the output generation step, and if it is determined that there is an inconsistency, re-executing the question provision step to provide an additional question for acquiring information necessary to resolve the inconsistency, and re-executing the answer acquisition step to acquire an additional answer;
5. The information processing method according to any one of appendices 1 to 4.

(Appendix 6)
a process of determining dependencies between the generated plurality of requirements and acquiring information about the dependencies;
6. An information processing method according to any one of appendices 1 to 5.

(Appendix 7)
a process of storing the requirements in a storage unit, retrieving the requirements from the storage unit at a predetermined timing, and providing a question including confirmation of whether the requirements have been updated,
7. An information processing method according to any one of appendices 1 to 6.

(Appendix 8)
The method includes a process of setting a predetermined requirement from among the plurality of requirements as a standard requirement, and a standard determination process of determining at least one of whether information on the other requirements is insufficient, whether there is an inconsistency, and whether there is an update, based on the standard requirement.
An information processing method according to any one of appendices 1 to 7.

(Appendix 9)
and a process of providing questions to resolve at least one of the cases where information on other requirements is insufficient, inconsistent, or needs to be updated based on the determination result of the criteria determination process.
9. The information processing method according to claim 8.

(Appendix 10)
a process of acquiring a deletion instruction for the request item, and a process of deleting the deletion instruction when the deletion instruction is acquired,
10. The information processing method according to any one of appendices 1 to 9.

(Appendix 11)
A process of generating a document in which the requirements are classified according to a predetermined standard;
and inputting the requirements into the trained model for each classification of the document to output secondary requirements.
An information processing method according to any one of appendices 1 to 10.

(Appendix 12)
and converting the secondary requirements into a document organized in accordance with a predetermined format.
12. The information processing method according to claim 11.

(Appendix 13)
In the information processing device,
a question providing step of performing at least one of selecting a question for obtaining requirements and generating the question;
an answer acquisition step of acquiring an answer to the question;
an output generation step of generating the requirements from the answers using a trained model that has learned at least text data and generates requirements estimated from the answers when information including the answers is input;
A program for executing an information processing method including the steps of:

(Appendix 14)
An information processing system executed by an information processing device,
a question providing unit that performs at least one of selecting a question for acquiring requirements and generating the question;
an answer acquisition unit that acquires an answer to the question;
an output generation unit that generates the requirements from the answers using a trained model that has learned at least text data and that generates requirements estimated from the answers when information including the answers is input;
An information processing system having the above.

The embodiments disclosed herein are illustrative in all respects and should not be considered limiting. The scope of the present invention is defined by the claims, not by the meaning described above, and is intended to include all modifications equivalent to the meaning and scope of the claims. Furthermore, the present invention is not limited to the above-described embodiments; various modifications are possible within the scope of the claims, and embodiments obtained by appropriately combining the technical means disclosed in different embodiments are also included in the technical scope of the present invention.

10 Information processing system 12 Server (information processing device)
22 Storage unit 40 Information processing system 50 Server (information processing device)
60 Storage unit 80 Information processing system 82 Server (information processing device)
220 Program 222 Trained model 242 Question providing unit 244 Information acquisition unit (answer acquisition unit)
246 Information processing unit (output generation unit)
600 Program 700 Question providing unit 702 Information acquisition unit (answer acquisition unit)
704 Information processing unit (output generation unit)
900 Program 1002 Information Acquisition Unit (Response Acquisition Unit)
1004 Information processing unit (output generation unit)

Claims (16)

An information processing method executed by an information processing device,
a question providing step of performing at least one of selecting a question for obtaining requirements and generating the question;
an answer acquisition step of acquiring an answer to the question;
an output generation step of generating the requirements from the answers using a trained model that has been trained on at least text data and that generates requirements estimated from the answers when information including the answers is input;
Including,
determining whether or not there is an inconsistency among the plurality of requirements generated in the output generation step, and if it is determined that there is an inconsistency, re-executing the question provision step to provide an additional question for acquiring information necessary to resolve the inconsistency, and re-executing the answer acquisition step to acquire an additional answer;
Information processing methods. An information processing method executed by an information processing device,
a question providing step of performing at least one of selecting a question for obtaining requirements and generating the question;
an answer acquisition step of acquiring an answer to the question;
an output generation step of generating the requirements from the answers using a trained model that has been trained on at least text data and that generates requirements estimated from the answers when information including the answers is input;
Including,
a process of determining dependencies between the generated plurality of requirements and acquiring information about the dependencies;
Information processing methods. An information processing method executed by an information processing device,
a question providing step of performing at least one of selecting a question for obtaining requirements and generating the question;
an answer acquisition step of acquiring an answer to the question;
an output generation step of generating the requirements from the answers using a trained model that has learned at least text data and generates requirements estimated from the answers when information including the answers is input;
Including,
The method includes a process of setting a predetermined requirement from among the plurality of requirements as a standard requirement, and a standard determination process of determining at least one of whether information on the other requirements is insufficient, whether there is an inconsistency, and whether there is an update, based on the standard requirement.
Information processing methods. the step of providing a question includes a process of providing background information on the question together with the question;
The information processing method according to any one of claims 1 to 3 . the output generation step includes a process of generating at least one of an evaluation of the answer based on predetermined evaluation criteria and a reason for the evaluation;
The information processing method according to any one of claims 1 to 3 . a verification step of analyzing the requirements generated by the output generation step and detecting missing information in at least one of the questions and the answers;
When a lack of information is detected in the verification step, the question providing step is executed again to provide an additional question for acquiring the lacking information, and the answer obtaining step is executed again to obtain the additional answer.
The information processing method according to any one of claims 1 to 3 . determining whether or not there is an inconsistency among the plurality of requirements generated in the output generation step, and if it is determined that there is an inconsistency, re-executing the question provision step to provide an additional question for acquiring information necessary to resolve the inconsistency, and re-executing the answer acquisition step to acquire an additional answer;
4. The information processing method according to claim 2 or 3 . a process of determining dependencies between the generated plurality of requirements and acquiring information about the dependencies;
4. The information processing method according to claim 1 or 3 . a process of storing the requirements in a storage unit, retrieving the requirements from the storage unit at a predetermined timing, and providing a question including confirmation of whether the requirements have been updated,
The information processing method according to any one of claims 1 to 3 . The method includes a process of setting a predetermined requirement from among the plurality of requirements as a standard requirement, and a standard determination process of determining at least one of whether information on the other requirements is insufficient, whether there is an inconsistency, and whether there is an update, based on the standard requirement.
3. The information processing method according to claim 1 . and a process of providing questions to resolve at least one of the cases where information on other requirements is insufficient, inconsistent, or needs to be updated based on the determination result of the criteria determination process.
The information processing method according to claim 10 . a process of acquiring a deletion instruction for the request item, and a process of deleting the deletion instruction when the deletion instruction is acquired,
The information processing method according to any one of claims 1 to 3 . A process of generating a document in which the requirements are classified according to a predetermined standard;
and inputting the requirements into the trained model for each classification of the document to output secondary requirements.
The information processing method according to any one of claims 1 to 3 . and converting the secondary requirements into a document organized in accordance with a predetermined format.
The information processing method according to claim 13 . In the information processing device,
a question providing step of performing at least one of selecting a question for obtaining requirements and generating the question;
an answer acquisition step of acquiring an answer to the question;
an output generation step of generating the requirements from the answers using a trained model that has learned at least text data and generates requirements estimated from the answers when information including the answers is input;
Including,
determining whether or not there is an inconsistency among the plurality of requirements generated in the output generation step, and if it is determined that there is an inconsistency, re-executing the question provision step to provide an additional question for acquiring information necessary to resolve the inconsistency, and re-executing the answer acquisition step to acquire an additional answer;
A program for executing an information processing method. An information processing system including an information processing device,
a question providing unit that performs at least one of selecting a question for acquiring requirements and generating the question;
an answer acquisition unit that acquires an answer to the question;
an output generation unit that generates the requirements from the answers using a trained model that has learned at least text data and that generates requirements estimated from the answers when information including the answers is input;
and
determining whether or not there is an inconsistency among the plurality of requirements generated by the output generation unit, and if it is determined that there is an inconsistency, the question provision unit provides an additional question for acquiring information necessary to resolve the inconsistency again, and the answer acquisition unit executes a process of acquiring an additional answer again;
Information processing system.