Systems, computers, and data manufacturing methods

Abstract

The goal is to provide a system that can assist in the creation of other systems. [Solution] In this invention, the storage device of one or more computers stores first information relating to one or more first functions provided by the software platform, and second information relating to one or more second functions provided by each of the one or more plugins available in the software platform. The system plan data indicates whether one or more functions required by the generation system can be realized by the first function, second function, or third function. The control device of one or more computers generates third information related to the input data from the first information and the second information, and causes a machine learning model to generate system plan data for the generation system based on the input data and the third information.

Description

Technical Field

[0001] The present invention relates to a system, a computer, and a data manufacturing method.

Background Art

[0002] As an invention related to a conventional system, for example, a platform described in Non-Patent Document 1 is known. This platform can create, for example, a business system for customer management. Also, a plurality of plugins are prepared in the platform. By using the plugins, the user can expand the functions of the platform.

Prior Art Documents

Non-Patent Documents

[0003]

Non-Patent Document 1

Summary of the Invention

Problems to be Solved by the Invention

[0004] By the way, a SIer (System Integrator) undertakes a series of processes such as the design, development, introduction, operation, and maintenance of information systems used in a company. When introducing an information system using the above platform, the SIer examines which plugins to use and the necessity of developing a new program after understanding the functions of the above platform.

[0005] Therefore, an object of the present invention is to provide a system, a computer, and a data manufacturing method that can support the creation of a system.

Means for Solving the Problems

[0006] The first form is, A system for generating system planning document data used in a software platform that supports the creation of generation systems, The aforementioned system comprises one or more computers. The storage devices of the one or more computers store first information relating to one or more first functions provided by the software platform, and second information relating to one or more second functions provided by each of the one or more plugins available in the software platform. The software platform can add the first function to the generation system, The plugin can add the second function to the generation system. The system planning document data indicates whether the one or more functions required by the generation system can be realized by one or more first functions, one or more second functions, or one or more first functions and a third function that is not one or more second functions. The aforementioned one or more computers are equipped with machine learning models, The control devices of the one or more computers described above are: Get input data that includes text, A third piece of information related to the input data is generated from the first piece of information and the second piece of information. Based on the input data and the third information, the machine learning model generates the system planning document data for the generation system. It is a system.

[0007] The second form is, The control devices of the one or more computers described above are: Based on audio data representing the audio in the meeting, the input data is generated by converting the audio into text. In the process of acquiring the aforementioned input data, the input data obtained is obtained in which the audio has been converted into text. This is the system described in the first form.

[0008] The third form is, The control devices of the one or more computers described above are: In response to the acquisition of the audio data during the aforementioned meeting, the input data is generated by sequentially converting the audio into text. This is the system described in the second form.

[0009] The fourth form is, The control devices of the one or more computers described above are: Based on the aforementioned audio data, emotional information regarding the emotions of the speakers in the meeting is generated. Based on the aforementioned audio data and the aforementioned emotion information, the input data is generated in which the audio is converted into text. This is the system described in the second or third form.

[0010] The fifth form is, The control devices of the one or more computers described above are: Based on the system planning document data, overall explanatory data is generated that describes the system configuration of the generation system realized by at least one of the first function, the second function, and the third function. The system is one of the first to fourth forms.

[0011] The sixth form is, The control devices of the one or more computers described above are: Based on the system planning document data, the machine learning model generates second explanatory data describing the plugin for the second function that can realize one or more functions required for the generation system. The system is one of the first to fifth forms.

[0012] The seventh form is, The control devices of the one or more computers described above are: Based on the system planning document data, causing the machine learning model to generate third explanatory data for explaining one or more functions required for the generation system that can be realized by the third function. The system according to any one of the first to sixth forms.

[0013] The eighth form is The third explanatory data includes an explanation of a generation application that realizes the third function. The system according to the seventh form.

[0014] The ninth form is The machine learning model is a trained model that performs machine learning using data related to the data output by the machine learning model as teacher data. The system according to any one of the first to eighth forms.

[0015] The tenth form is A first computer used in a system for generating system planning document data used in a software platform for assisting in the creation of a generation system, The system includes the first computer and a second computer. The storage device of the first computer stores first information regarding one or more first functions provided by the software platform and second information regarding one or more second functions provided by each of one or more plugins available in the software platform. The software platform can add the first function to the generation system. The plugin can add the second function to the generation system. The system planning document data indicates which of the one or more first functions, the one or more second functions, or a third function that is not the one or more first functions and the one or more second functions can realize one or more functions required for the generation system. The second computer includes a machine learning model. The control unit of the first computer described above is Get input data that includes text, A third piece of information related to the input data is generated from the first piece of information and the second piece of information. Based on the input data and the third information, the second computer is instructed to generate system planning document data for the generation system in the machine learning model. It is the first computer.

[0016] The 11th form is, A second computer used in a system that generates system planning document data used in a software platform that supports the creation of generation systems, The aforementioned system comprises a first computer and a second computer. The storage device of the first computer stores first information relating to one or more first functions provided by the software platform, and second information relating to one or more second functions provided by each of the one or more plug-ins available in the software platform. The software platform can add the first function to the generation system, The plugin can add the second function to the generation system. The system planning document data indicates whether the one or more functions required by the generation system can be realized by one or more first functions, one or more second functions, or one or more first functions and a third function that is not one or more second functions. The aforementioned second computer is equipped with a machine learning model, The control unit of the first computer described above is Get input data that includes text, A third piece of information related to the input data is generated from the first piece of information and the second piece of information. The control unit of the second computer is Based on the input data and the third information, the machine learning model generates the system planning document data for the generation system. This is the second computer.

[0017] The 12th form is, A data manufacturing method for generating system planning document data used in a software platform that supports the creation of a generation system, The system consists of one or more computers. The storage devices of the one or more computers store first information relating to one or more first functions provided by the software platform, and second information relating to one or more second functions provided by each of the one or more plugins available in the software platform. The software platform can add the first function to the generation system, The plugin can add the second function to the generation system. The system planning document data indicates whether the one or more functions required by the generation system can be realized by one or more first functions, one or more second functions, or one or more first functions and a third function that is not one or more second functions. The aforementioned one or more computers are equipped with machine learning models, The control device of the one or more computers: Get input data that includes text, From the first and second information, a third piece of information related to the input data is generated. Based on the input data and the third information, the machine learning model generates the system planning document data for the generation system. This is a data creation method. [Effects of the Invention]

[0018] According to the present invention, it is possible to support the creation of a system. [Brief explanation of the drawing]

[0019] [Figure 1] Figure 1 is an explanatory diagram of System 1. [Figure 2] Figure 2 is a block diagram of user terminal 10. [Figure 3] Figure 3 is a block diagram of server 110. [Figure 4] Figure 4 is a block diagram of server 210. [Figure 5] Figure 5 shows the meeting minutes data D2. [Figure 6] Figure 6 shows the standard function table T1, which is the first piece of information. [Figure 7] Figure 7 shows the second piece of information, the plug-in table T2. [Figure 8] Figures 8 through 10 show the system planning document data D3. [Figure 9] Figures 8 through 10 show the system planning document data D3. [Figure 10] Figures 8 through 10 show the system planning document data D3. [Figure 11] Figure 11 is a flowchart showing the actions performed by the control device 12 of the user terminal 10. [Figure 12] Figure 12 is a flowchart showing the actions performed by the control device 112 of server 110 and the control device 212 of server 210 in system 1. [Figure 13] Figure 13 is a diagram of the functional requirements document shown by functional requirements document data D5. [Figure 14] Figure 14 is a diagram of the non-functional requirements document shown in non-functional requirements document data D6. [Figure 15] Figure 15 is the business process diagram shown in business process diagram data D7. [Figure 16] Figure 16 is a diagram of the application design document shown in application design document data D8. [Figure 17] Figure 17 is a diagram of the plugin proposal shown in plugin proposal data D9. [Figure 18]Figure 18 is a diagram of the management item sheet shown in management item sheet data D10. [Figure 19] Figure 19 is the ER diagram shown for ER diagram data D11. [Figure 20] Figure 20 is the system configuration diagram shown in system configuration diagram data D12. [Modes for carrying out the invention]

[0020] (Embodiment) System 1 according to an embodiment of this disclosure will be described with reference to the drawings.

[0021] [System 1 Configuration] First, the overall configuration of System 1 will be explained with reference to the diagrams. Figure 1 is an explanatory diagram of System 1. Figure 2 is a block diagram of user terminal 10. Figure 3 is a block diagram of server 110. Figure 4 is a block diagram of server 210.

[0022] System 1, shown in Figure 1, comprises a user terminal 10 (one or more computers), a server 110 (one or more computers), and a server 210 (one or more computers). The user terminal 10, server 110, and server 210 can communicate with each other via a communication network. This network could be the internet, an intranet, or similar.

[0023] The user terminal 10 is an information processing device used by a user. The user terminal 10 is, for example, a smartphone, a tablet device, or a personal computer. As shown in Figure 2, the user terminal 10 includes a control unit 12, a storage device 14, a network interface 16, a graphics processing unit 18, a display 20, an operation unit 26, and a touch panel 28.

[0024] The storage device 14 stores programs and data. The storage device 14 is, for example, a combination of ROM (Read Only Memory), RAM (Random Access Memory), and storage (for example, flash memory or hard disk).

[0025] The program includes, for example, the following: • OS (Operating System) programs • Programs for applications that perform information processing (e.g., web browsers, or target applications described later)

[0026] The data includes, for example, the following: • Databases referenced in information processing • Data obtained by performing information processing (i.e., the results of performing information processing)

[0027] The control device 12 implements the functions of the user terminal 10 by executing a program stored in the storage device 14. The control device 12 is, for example, at least one of the following: ·CPU(Central Processing Unit) ·GPU(Graphic Processing Unit) ·ASIC(Application Specific Integrated Circuit) ·FPGA(Field Programmable Array)

[0028] The control device 12 includes a text conversion means 30 as a functional block.

[0029] The network interface 16 controls communication between the user terminal 10 and external devices. The external devices are servers 110 and 210.

[0030] The graphics processing unit 18 displays an image on the display 20 based on the image data generated by the control device 12. The display 20 is either a liquid crystal display or an organic EL (Electro-Luminescence) display.

[0031] The operation unit 26 generates an operation signal based on the user's operation via the touch panel 28 and outputs the operation signal to the control device 12.

[0032] Server 110 is the first computer. Server 110 is a RAG (Retrieval-Augmented Generation) server. Therefore, based on a query from a user, Server 110 retrieves information related to the query from a database and knowledge base. Then, based on the query and the information related to the query, Server 110 generates a context and outputs the context to Server 210. As shown in Figure 3, Server 110 includes a control unit 112, a storage device 114, and a network interface 116.

[0033] The storage device 114 stores programs and data. The storage device 114 is, for example, a combination of ROM (Read Only Memory), RAM (Random Access Memory), and storage (for example, flash memory or hard disk).

[0034] The control device 112 implements the functions of the server 110 by executing the program stored in the storage device 114. The control device 112 is, for example, at least one of the following: ·CPU(Central Processing Unit) ·GPU(Graphic Processing Unit) ·ASIC(Application Specific Integrated Circuit) ·FPGA(Field Programmable Array)

[0035] The control device 112 includes, as a functional block, an input data acquisition means 120, a third information generation means 122, and a system planning document data generation instruction means 124.

[0036] The network interface 116 controls communication between the server 110 and external devices. The external devices are the user terminal 10 and the server 210.

[0037] Server 210 is a second computer. Server 210 is a so-called machine learning server. Server 210 generates answers based on the context generated by Server 110. As shown in Figure 4, Server 210 includes a control unit 212, a storage device 214, and a network interface 216.

[0038] The storage device 214 stores programs and data. The storage device 214 is, for example, a combination of ROM (Read Only Memory), RAM (Random Access Memory), and storage (for example, flash memory or hard disk).

[0039] The control device 212 implements the functions of the server 210 by executing the program stored in the storage device 214. The control device 212 is, for example, at least one of the following: ·CPU(Central Processing Unit) ·GPU(Graphic Processing Unit) ·ASIC(Application Specific Integrated Circuit) ·FPGA(Field Programmable Array)

[0040] The control device 212 includes, as functional blocks, a system planning document data generation means 220, an overall explanation data generation means 222, a second explanation data generation means 226, and a third explanation data generation means 228.

[0041] The network interface 216 controls communication between the server 210 and external devices. The external devices are the user terminal 10 and the server 110.

[0042] Furthermore, the server 210 (one or more computers) is equipped with a machine learning model 250. The machine learning program is a program that executes a machine learning algorithm to find certain rules from the training data and generate a trained machine learning model 250 that represents the found rules. When the control device 212 of the server 210 executes the machine learning program, multiple training data are trained and the parameters of the inference program are adjusted. As a result, a trained machine learning model 250 is generated. The machine learning model 250 is a trained model that has been trained using data related to the data output by the machine learning model 250 as training data. The data related to the data output by the machine learning model 250 is, for example, some or all of the following: meeting minutes data D2, system planning document data D3, functional requirements document data D5, non-functional requirements document data D6, business flow diagram data D7, application design document data D8, plugin proposal document data D9, management item document data D10, ER diagram data D11, and system configuration diagram data D12.

[0043] The machine learning algorithm is not particularly limited as long as it is supervised learning, and may include, for example, decision trees, nearest neighbors, Naive Bayesian classifiers, support vector machines, or neural networks. Therefore, the trained machine learning model 250 includes decision trees, nearest neighbors, Naive Bayesian classifiers, support vector machines, or neural networks. Backpropagation may be used in the machine learning process that generates the trained machine learning model 250.

[0044] For example, a neural network includes an input layer, one or more hidden layers, and an output layer. Specifically, neural networks include deep neural networks, recurrent neural networks, or convolutional neural networks, and perform deep learning. A deep neural network, for example, includes an input layer, multiple hidden layers, and an output layer.

[0045] [System 1 Operation] Next, the operation of System 1 will be explained with reference to the diagrams. Figure 5 shows the meeting minutes data D2. Figure 6 shows the standard function table T1, which is the first information. Figure 7 shows the plug-in table T2, which is the second information. Figures 8 through 10 show the system planning document data D3.

[0046] First, let's explain the operation overview of System 1. System 1 generates system planning document data D3, which is used in platform K (software platform) that supports the creation of System A (generation system). More specifically, the user is the developer of System A. The user holds a meeting with a client to discuss the creation of System A. During this meeting, System 1 acquires video data D1 of the meeting. Based on the video data D1, System 1 generates meeting minutes data D2, as shown in Figure 5. Meeting minutes data D2 is the minutes of the meeting about System A. Meeting minutes data D2 describes the requirements for System A. Specifically, meeting minutes data D2 describes how System A will be implemented and the functions that should be implemented in System A.

[0047] Here, we will explain System A. System A is used by the secretariat of the "Umeda Yukata Festival". System A handles the sending, collection, tabulation, and analysis of questionnaires, as well as contacting questionnaire respondents. System A is created using Platform K. Platform K can be used to create System A for customer management. Platform K is a no-code application that allows users to create systems using a visual-based interface. Platform K is, for example, kintone® manufactured by Cybozu, Inc.

[0048] Platform K can implement one or more first functions and one or more second functions. One or more first functions are standard functions provided by Platform K (software platform). Platform K (software platform) can add first functions to System A (generation system). First functions include, for example, the processing of collecting questionnaires. Second functions are functions provided by one or more plugins available on Platform K (software platform). Plugins can add second functions to System A (generation system). Second functions include, for example, the processing of contacting questionnaire respondents. Furthermore, functions that are neither one or more first functions nor one or more second functions are defined as third functions. Third functions include, for example, the processing of creating questionnaire response forms and the processing of sending questionnaire forms via email, fax, etc.

[0049] The storage device 114 of server 110 (one or more computers) stores first information relating to one or more first functions and second information relating to one or more second functions. More specifically, the storage device of server 110 stores the standard function table T1 shown in Figure 6 and the plug-in table T2 shown in Figure 7. The standard function table T1 is the first information. The standard function table T1 contains a description of the first functions that can be realized by the standard functions provided by platform K. The plug-in table T2 is the second information. The plug-in table T2 contains a description of the second functions that can be realized by plug-ins available for platform K.

[0050] System 1 causes the machine learning model 250 to generate system planning data D3, shown in Figures 8 to 10, based on meeting minutes data D2. System planning data D3 indicates whether one or more functions required for System A (generation system) can be realized by one or more first functions, one or more second functions, or a third function. Specifically, system planning data D3 includes a use case table 300. The user can design and develop System A based on this system planning data D3.

[0051] Next, the operation of System 1 will be explained in detail with reference to the diagrams. Figure 11 is a flowchart of the operations performed by the control device 12 of the user terminal 10. Figure 12 is a flowchart of the operations performed by the control device 112 of server 110 and the control device 212 of server 210 in System 1.

[0052] The control device 12 of the user terminal 10 reads the programs stored in the storage device 14, and these programs cause the control device 12 of the user terminal 10 to execute the operations described below. The programs cause the control device 12 to execute the text conversion means 30.

[0053] The control device 112 of the server 110 reads the programs stored in the storage device 114, and these programs cause the control device 112 of the server 110 to execute the operations described below. The programs cause the control device 112 to execute the input data acquisition means 120, the third information generation means 122, and the system planning document data generation instruction means 124.

[0054] The control device 212 of the server 210 reads the programs stored in the memory device 214, causing these programs to perform the operations described below on the control device 212 of the server 210. The programs cause the control device 212 to execute the system planning data generation means 220, the overall explanation data generation means 222, the second explanation data generation means 226, and the third explanation data generation means 228.

[0055] First, the user conducts a web conference with the client. A web conferencing server (not shown) generates and stores the video data D1 of the web conference. In parallel with the generation and storage of the video data D1 by the web conferencing server, the control device 12 of the user terminal 10 downloads the video data D1 from the web conferencing server. As a result, the control device 12 of the user terminal 10 acquires the video data D1 (step S201).

[0056] Furthermore, the control device 12 (text conversion means 30) of the user terminal 10 generates meeting minutes data D2 (input data - see Figure 5) in which the audio has been converted to text based on the video data D1 from the meeting (step S202). In this embodiment, the control device 12 (text conversion means 30) of the user terminal 10 generates meeting minutes data D2 (input data) by sequentially converting the audio to text in response to the acquisition of video data D1 (audio data) during the meeting. Since the video data D1 includes audio, it corresponds to audio data that represents audio. The processing in step S202 can be implemented, for example, by Google Cloud's Speech-to-Text (manufactured by Google). Sequential conversion means that the control device 12 of the user terminal 10 converts the audio to text while acquiring the video data D1. In sequential conversion, the control device 12 of the user terminal 10 operates in such a way that the delay from the acquisition of video data D1 to the conversion of audio to text is minimized as much as possible.

[0057] Next, the control device 12 of the user terminal 10 transmits the meeting minutes data D2 to the server 110 via the network interface 16 (step S203). Accordingly, the network interface 116 of the server 110 receives the meeting minutes data D2 and outputs the meeting minutes data D2 to the control device 112. As a result, the control device 112 (input data acquisition means 120) of the server 110 (one or more computers) acquires the meeting minutes data D2 (input data) in which the speech has been converted to text (step S101). In this way, the control device 112 (input data acquisition means 120) of the server 110 (one or more computers) acquires the meeting minutes data D2 (input data) which includes text as shown in Figure 5.

[0058] Next, the control device 112 (third information generation means 122) of the server 110 generates a context D4 (third information) related to the meeting minutes data D2 (input data) from the standard function table T1 (first information) shown in Figure 6 and the plug-in table T2 (second information) shown in Figure 7 (step S102). The context D4 includes the results of the server 110's control device 112 retrieving information about the meeting minutes data D2 based on the standard function table T1 and the plug-in table T2.

[0059] Next, the control device 112 of server 110 transmits the meeting minutes data D2 and context D4 to server 210 via the network interface 116 (step S103). That is, in step S103, the control device 112 of server 110 (system plan data generation instruction means 124) instructs server 210 (second computer) to generate system plan data D3 (see Figures 8 to 10) for system A (generation system) in the machine learning model 250 based on the meeting minutes data D2 (input data) and context D4 (third information). Accordingly, the network interface 216 of server 210 receives the meeting minutes data D2 and context D4 and outputs the meeting minutes data D2 and context D4 to the control device 212. As a result, the control device 212 of server 210 acquires the meeting minutes data D2 and context D4 (step S1).

[0060] Next, the control device 212 (system planning document data generation means 220) of the server 210 causes the machine learning model 250 to generate system planning document data D3 for system A (generating system) based on the meeting minutes data D2 (input data) and context D4 (third information) (step S2). As shown in Figures 8 to 10, the system planning document data D3 shows the specifications of system A described in the meeting minutes data D2. Specifically, the system planning document data D3 includes the purpose, main functions 303, system configuration, benefits, other information 302, business flow 301, use case table 300, and supplementary information 304.

[0061] The "Main Functions" column (303) lists the functions required of System 1. The "Other" column (302) describes information other than the functions required of System A. The "Business Flow" column (301) describes the tasks performed by those involved with System A (secretariat, respondents, and system).

[0062] Use Case Table 300 is a table that shows whether one or more functions required by System A (Generation System) can be realized by one or more first functions, one or more second functions, or a third function. The Use Case column describes the function. The Classification column describes whether the function in the Use Case column can be realized by a first function, second function, or third function. Standard functions correspond to first functions. Plugins correspond to second functions. Customization means that the user customizes the standard functions and corresponds to third functions. External services correspond to third functions.

[0063] The section labeled "Supplementary Information 304" contains supplementary information. In this embodiment, Supplementary Information 304 describes the second and third functions that cannot be realized by the standard functions (first function) of platform K.

[0064] Next, the control device 212 of the server 210 causes the machine learning model 250 to generate functional requirements data D5 based on the system planning data D3 (step S3). Figure 13 is a diagram of the functional requirements document shown by the functional requirements data D5. As shown in Figure 13, the functional requirements document corresponds to the use case table 300 of the system planning data D3.

[0065] Next, the control device 212 of the server 210 causes the machine learning model 250 to generate non-functional requirements data D6 based on the system planning data D3 (step S4). Figure 14 is a diagram of the non-functional requirements shown in the non-functional requirements data D6. As shown in Figure 14, the non-functional requirements document describes information other than the functions required for system A. The non-functional requirements document describes the number of users and data records expected for system A. The non-functional requirements document corresponds to "Other" 302 in the system planning data D3.

[0066] Next, the control device 212 of the server 210 causes the machine learning model 250 to generate business flow diagram data D7 based on the system planning document data D3 (step S5). Figure 15 is the business flow diagram shown by the business flow diagram data D7. As shown in Figure 15, the business flow diagram illustrates the business flow of the system planning document data D3. In this way, the control device 212 of the server 210 (overall explanation data generation means 222) generates overall explanation data that describes the system configuration of system A (generated system) realized by at least one of the first function, second function and third function, based on the system planning document data D3. In this embodiment, the overall explanation data is the business flow diagram data D7. The business flow diagram corresponds to the business flow 301 of the system planning document data D3.

[0067] Next, the control device 212 (third explanatory data generation means 228) of server 210 causes the machine learning model 250 to generate application design data D8 (third explanatory data) that describes one or more functions required for system A (generation system) that can be realized by the third function, based on the system planning data D3 (step S6). In step S6, the control device 112 of server 110 generates a context based on the system planning data D3. The control device 212 of server 210 generates application design data D8 based on the system planning data D3 and the context. The application design data D8 (third explanatory data) includes a description of the generation application that realizes the third function. Figure 16 is a diagram of the application design shown in the application design data D8. As shown in Figure 16, the application design describes the functions required for the system to realize the third function. The application design corresponds to the use case table 300 and supplementary information 304 of the system planning data D3. Furthermore, the application design document also includes information on the first function that can be implemented using the standard functions of Platform K, and the second function that can be implemented using plugins available for Platform K.

[0068] Next, the control device 212 (second descriptive data generation means 226) of server 210 causes the machine learning model 250 to generate plugin proposal data D9 (second descriptive data) that describes plugins for second functions capable of realizing one or more functions required for system A (generation system), based on the system planning data D3 (more precisely, functional requirements data D5) (step S7). In step S7, the control device 112 of server 110 generates a context based on the system planning data D3 (functional requirements data D5). The control device 212 of server 210 generates plugin proposal data D9 based on the system planning data D3 (functional requirements data D5) and the context. Figure 17 is a diagram of the plugin proposal shown in the plugin proposal data D9. As shown in Figure 17, the plugin proposal describes a list of plugins having second functions that are described as being achievable by plugins in the functional requirements shown in Figure 13. The plugin proposal corresponds to the use case table 300 and supplementary information 304 of the system planning data D3.

[0069] Next, the control unit 212 of server 210 causes the machine learning model 250 to generate management item data D10 based on the application design data D8 and the plugin proposal data D9 (step S8). In step S8, the control unit 112 of server 110 generates a context based on the application design data D8 and the plugin proposal data D9. The control unit 212 of server 210 generates management item data D10 based on the application design data D8, the plugin proposal data D9 and the context. Figure 18 is a diagram of the management item document shown in the management item data D10. As shown in Figure 18, the management item document lists details of the information managed in the system for implementing the standard function to realize the first function, the plugin to realize the second function, and the system to realize the third function.

[0070] Next, the control device 212 of the server 210 causes the machine learning model 250 to generate ER (Entry Relationship) diagram data D11 based on the management item data D10 (step S9). Figure 19 is the ER diagram shown by the ER diagram data D11. As shown in Figure 19, the ER diagram describes the structure of the data exchanged in system A. In this embodiment, the ER diagram describes the structure of the data exchanged in sending and receiving questionnaires and sending and receiving responses.

[0071] Next, the control device 212 of the server 210 causes the machine learning model 250 to generate system configuration diagram data D12 based on the application design document data D8 (step S10). Figure 20 is the system configuration diagram shown by the system configuration diagram data D12. As shown in Figure 20, the system configuration diagram is a diagram that visually represents the components and their interrelationships in a system in which system A is used. Specifically, the components are the administrator (user), system A, facility, and parent facility. The system configuration diagram also describes the data exchange between the administrator (user) and system A. In this way, the control device 212 of the server 210 (overall explanation data generation means 222) generates overall explanation data that describes the system configuration of system A (generated system) realized by at least one of the first function, second function, and third function, based on the system planning document data D3. In this embodiment, the overall explanation data is the system configuration diagram data D12.

[0072] Next, the control device 212 of server 210 transmits system planning document data D3, functional requirements document data D5, non-functional requirements document data D6, business flow diagram data D7, application design document data D8, plug-in proposal document data D9, management item document data D10, ER diagram data D11, and system configuration diagram data D12 to server 110 via the network interface 216 (step S11). Accordingly, the network interface 116 of server 110 receives system planning document data D3, functional requirements document data D5, non-functional requirements document data D6, business process flow diagram data D7, application design document data D8, plug-in proposal document data D9, management item document data D10, ER diagram data D11, and system configuration diagram data D12, and outputs system planning document data D3, functional requirements document data D5, non-functional requirements document data D6, business process flow diagram data D7, application design document data D8, plug-in proposal document data D9, management item document data D10, ER diagram data D11, and system configuration diagram data D12 to the control device 112. As a result, the control device 112 of server 110 acquires system planning document data D3, functional requirements document data D5, non-functional requirements document data D6, business process flow diagram data D7, application design document data D8, plug-in proposal document data D9, management item document data D10, ER diagram data D11, and system configuration diagram data D12 (step S104).

[0073] Next, the control device 112 of the server 110 transmits the system planning document data D3, functional requirements document data D5, non-functional requirements document data D6, business flow diagram data D7, application design document data D8, plug-in proposal document data D9, management item document data D10, ER diagram data D11, and system configuration diagram data D12 to the user terminal 10 via the network interface 116 (step S105). Accordingly, the user terminal 10 receives the system planning document data D3, functional requirements document data D5, non-functional requirements document data D6, business flow diagram data D7, application design document data D8, plug-in proposal document data D9, management item document data D10, ER diagram data D11, and system configuration diagram data D12. As a result, by operating the user terminal 10, the user can view the system planning document data D3, functional requirements document data D5, non-functional requirements document data D6, business process flow diagram data D7, application design document data D8, plugin proposal document data D9, management item document data D10, ER diagram data D11, and system configuration diagram data D12.

[0074] [effect] System 1 can assist in the creation of System A. More specifically, System 1 aims to generate system planning data D3. System planning data D3 indicates whether the one or more functions required for System A can be realized by one or more first functions, one or more second functions, or one or more first functions and one or more second functions that are not third functions. Therefore, the storage device 114 of Server 110 stores a standard function table T1 relating to the one or more first functions provided by Platform K, and a plugin table T2 relating to the one or more second functions provided by each of the one or more plugins available on Platform K. This allows the control device 112 of Server 110 to determine whether the function required for System A in the minutes data D2 corresponds to a first function, a second function, or a third function. Therefore, the control device 112 of Server 110 generates a context D4 related to the minutes data D2 from the standard function table T1 and the plugin table T2. Context D4 contains information for determining whether the function required for System A in the meeting minutes data D2 corresponds to the first function, second function, or third function. Based on this, the control device 212 of server 210 can generate the system plan data D3 for System A in the machine learning model 250 based on the meeting minutes data D2 and context D4. Thus, since the user can create System A using the system plan data D3, System 1 can assist in the creation of System A.

[0075] In System 1, the control device 12 of the user terminal 10 generates meeting minutes data D2, in which the audio is converted to text, based on video data D1 showing the audio from the meeting. This allows the user to easily obtain the meeting minutes data D2, which includes the text.

[0076] In System 1, the control device 12 of the user terminal 10 generates meeting minutes data D2 by sequentially converting audio to text in response to the acquisition of video data D1 during the meeting. As a result, the user can obtain the meeting minutes data D2 at the end of the meeting.

[0077] System 1 can assist in the creation of System A. More specifically, the control device 212 of server 210 generates business flow diagram data D7 and system configuration diagram data D12, which describe the system configuration of System A realized by at least one of the first, second, and third functions, based on the system planning document data D3. Thus, in System 1, the user can obtain business flow diagram data D7 and system configuration diagram data D12 in addition to the system planning document data D3. Therefore, the user can design and develop System A based on the system planning document data D3, business flow diagram data D7, and system configuration diagram data D12. Thus, System 1 can assist in the creation of System A.

[0078] System 1 can assist in the creation of System A. More specifically, the control device 212 of the server 210 causes the machine learning model 250 to generate plugin proposal data D9, which describes a second function plugin capable of realizing one or more functions required for System A, based on the system plan data D3. As a result, in System 1, the user can obtain the plugin proposal data D9 in addition to the system plan data D3. Therefore, the user can design and develop System A based on the system plan data D3 and the plugin proposal data D9. Thus, System 1 can assist in the creation of System A.

[0079] System 1 can assist in the creation of System A. More specifically, the control device 212 of the server 210 causes the machine learning model 250 to generate application design data D8, which describes one or more functions required for System A that can be realized by the third function, based on the system planning data D3. As a result, in System 1, the user can obtain the application design data D8 in addition to the system planning data D3. Therefore, the user can design and develop System A based on the system planning data D3 and the application design data D8. Thus, System 1 can assist in the creation of System A.

[0080] In System 1, the machine learning model 250 is a pre-trained model that has been trained using data related to the data output by the machine learning model 250 as training data. As a result, the machine learning model 250 learns each time System 1 generates system planning document data D3. Consequently, System 1 becomes able to generate system planning document data D3 that meets the user's requirements.

[0081] (Other embodiments) The system according to the present invention is not limited to System 1, but can be modified within the scope of its gist.

[0082] Furthermore, the control device 12 of the user terminal 10 may generate emotional information regarding the emotions of speakers in the meeting based on the video data D1 (audio data). The control device 12 of the user terminal 10 generates emotional information based on the content of the speakers' statements and their facial expressions. Then, the control device 12 of the user terminal 10 generates meeting minutes data D2 (input data) in which the audio has been converted to text, based on the video data D1 (audio data) and the emotional information. As a result, the emotions of the speakers in the meeting are reflected in the meeting minutes data D2. Consequently, the user can obtain meeting minutes data D2 that contains more information.

[0083] System 1 only needs to include one or more computers. Therefore, Server 110 and Server 210 may be implemented by a single computer. Also, User Terminal 10, Server 110, and Server 210 may be implemented by a single computer.

[0084] The meeting minutes data D2 may also be created by the user operating the user terminal 10. In this case, audio data is not required.

[0085] Furthermore, the control device 12 of the user terminal 10 may convert audio to text in a batch after the meeting has ended, rather than sequentially converting audio to text during the meeting.

[0086] Context D4 contains the results of a search for information related to meeting minutes data D2. However, context D4 may also contain information equivalent to meeting minutes data D2 in addition to the search results for information related to meeting minutes data D2. In this case, context D4 contains the third information and meeting minutes data D2.

[0087] Note that audio data may be used instead of video data D1.

[0088] The training data consists of part or all of the following: meeting minutes data D2, system planning document data D3, functional requirements document data D5, non-functional requirements document data D6, business process flow diagram data D7, application design document data D8, plugin proposal document data D9, management item document data D10, ER diagram data D11, and system configuration diagram data D12. However, the training data may also be modified versions of part or all of the following: meeting minutes data D2, system planning document data D3, functional requirements document data D5, non-functional requirements document data D6, business process flow diagram data D7, application design document data D8, plugin proposal data D9, management item document data D10, ER diagram data D11, and system configuration diagram data D12. Furthermore, the training data may also be context D4. [Explanation of symbols]

[0089] 1: System 10: User terminal 12: Control device 14:Storage device 16: Network Interface 18: Graphics Processing Unit 20: Display 26:Operation section 28: Touch panel 30: Text conversion method 110: Server 112: Control device 114: Storage device 116: Network Interface 120: Input data acquisition method 122:Third information generation means 124: System Planning Document Data Generation Instruction Method 210: Server 212: Control device 214: Storage device 216: Network Interface 220: System Planning Document Data Generation Method 222: Overall Description Data Generation Method 226: Second explanatory data generation means 228: Third explanatory data generation means 250: Machine Learning Models 300: Use Case Table A: System D1: Video data D2: Meeting minutes data D3: System Planning Document Data D4: Context D5: Functional Requirements Data D6: Non-functional requirements data D7: Business Flow Diagram Data D8: Application design document data D9: Plugin Proposal Data D10: Management item document data D11: ER diagram data D12: System Configuration Diagram Data K: Platform T1: Standard Function Table T2: Plugin Table

Claims

1. A system for generating system planning document data used in a software platform that supports the creation of generation systems, The aforementioned system comprises one or more computers. The storage devices of the one or more computers store first information relating to one or more first functions provided by the software platform, and second information relating to one or more second functions provided by each of the one or more plug-ins available in the software platform. The software platform can add the first function to the generation system, The plugin can add the second function to the generation system. The system planning document data indicates whether the one or more functions required by the generation system can be realized by one or more first functions, one or more second functions, or one or more first functions and a third function that is not one or more second functions. The aforementioned one or more computers are equipped with a machine learning model, The control devices of the one or more computers described above are: Get input data that includes text, A third piece of information related to the input data is generated from the first piece of information and the second piece of information. Based on the input data and the third information, the machine learning model generates the system planning document data for the generation system. system.

2. The control devices of the one or more computers described above are: Based on audio data representing the audio in the meeting, the input data is generated by converting the audio into text. In the process of acquiring the aforementioned input data, the input data obtained is obtained in which the audio has been converted into text. The system according to claim 1.

3. The control devices of the one or more computers described above are: In response to the acquisition of the audio data during the aforementioned meeting, the input data is generated by sequentially converting the audio into text. The system according to claim 2.

4. The control devices of the one or more computers described above are: Based on the aforementioned audio data, emotional information regarding the emotions of the speakers in the meeting is generated. Based on the aforementioned audio data and the aforementioned emotion information, the input data is generated in which the audio is converted into text. The system according to claim 2 or claim 3.

5. The control devices of the one or more computers described above are: Based on the system planning document data, overall explanatory data is generated that describes the system configuration of the generation system realized by at least one of the first function, the second function, and the third function. The system according to any one of claims 1 to 3.

6. The control devices of the one or more computers described above are: Based on the system planning document data, the machine learning model is instructed to generate second explanatory data describing the plugin for the second function that can realize one or more functions required for the generation system. The system according to any one of claims 1 to 3.

7. The control devices of the one or more computers described above are: Based on the system planning document data, the machine learning model is made to generate third explanatory data that describes one or more functions required for the generation system that can be realized by the third function. The system according to any one of claims 1 to 3.

8. The third explanatory data includes a description of the generation application that implements the third function. The system according to claim 7.

9. The aforementioned machine learning model is a trained model that has been trained using data related to the data output by the aforementioned machine learning model as training data. The system according to any one of claims 1 to 3.

10. A first computer used in a system that generates system planning document data used in a software platform that supports the creation of a generation system, The system comprises the first computer and the second computer. The storage device of the first computer stores first information relating to one or more first functions provided by the software platform, and second information relating to one or more second functions provided by each of the one or more plug-ins available in the software platform. The software platform can add the first function to the generation system, The plugin can add the second function to the generation system. The system planning document data indicates whether the one or more functions required by the generation system can be realized by one or more first functions, one or more second functions, or one or more first functions and a third function that is not one or more second functions. The aforementioned second computer is equipped with a machine learning model, The control device of the first computer is Get input data that includes text, A third piece of information related to the input data is generated from the first piece of information and the second piece of information. Based on the input data and the third information, the second computer is instructed to generate system planning document data for the generation system in the machine learning model. The first computer.

11. A second computer used in a system that generates system planning document data used in a software platform that supports the creation of generation systems, The aforementioned system comprises a first computer and a second computer. The storage device of the first computer stores first information relating to one or more first functions provided by the software platform, and second information relating to one or more second functions provided by each of the one or more plug-ins available in the software platform. The software platform can add the first function to the generation system, The plugin can add the second function to the generation system. The system planning document data indicates whether the one or more functions required by the generation system can be realized by one or more first functions, one or more second functions, or one or more first functions and a third function that is not one or more second functions. The aforementioned second computer is equipped with a machine learning model, The control device of the first computer is Get input data that includes text, A third piece of information related to the input data is generated from the first piece of information and the second piece of information. The control device of the second computer is Based on the input data and the third information, the machine learning model generates the system planning document data for the generation system. Second computer.

12. A data manufacturing method for generating system planning document data used in a software platform that supports the creation of a generation system, The system consists of one or more computers. The storage devices of the one or more computers store first information relating to one or more first functions provided by the software platform, and second information relating to one or more second functions provided by each of the one or more plug-ins available in the software platform. The software platform can add the first function to the generation system, The plugin can add the second function to the generation system. The system planning document data indicates whether the one or more functions required by the generation system can be realized by one or more first functions, one or more second functions, or one or more first functions and a third function that is not one or more second functions. The aforementioned one or more computers are equipped with a machine learning model, The control device of the one or more computers: Get input data that includes text, From the first and second information, a third piece of information related to the input data is generated. Based on the input data and the third information, the machine learning model generates the system planning document data for the generation system. Data creation methods.

Images