Information processing device, information processing method, and program

Abstract

To provide an information processing device, an information processing method, and a program, capable of facilitating converting of a system using a learned model into a WEB application.SOLUTION: An information processing device uses input information inputted to a user interface of a Web base by a user as input to a learned model, and executes processing using the learned model. The information processing device includes a definition information obtaining portion that obtains definition information defining the input information to be inputted to the learned model and output information outputted from the learned model, and a user interface generating portion for generating the interface, which includes an input area in which the user inputs the input information and an output area in which the output information is outputted, on the basis of the obtained definition information.SELECTED DRAWING: Figure 1

Description

[Technical field]

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

[0002] Conventionally, in the development of a system using AI (Artificial Intelligence) (hereinafter also referred to as an "AI system"), in addition to the development of an AI model, the development of a mechanism for using the AI ​​model is also required. As a mechanism for using an AI model (hereinafter also referred to as a "trained model"), for example, a development environment such as Jupyter Notebook can be mentioned. In the development environment, for example, by executing a program (hereinafter also referred to as an "execution program") that executes a process using the trained model, it is possible to use the developed trained model. For reference, Patent Document 1 listed below discloses a technology in which a management device generates a trained model, transmits the generated trained model to an edge terminal, and uses the trained model in the edge terminal.

[0003] However, it is difficult for general users to use the development environment. Therefore, in order to allow general users to easily use trained models, it is necessary to develop a user interface (UI) and data input / output functions as an application or web application. Web application development is an effective way to allow general users to widely and easily use systems that use trained models. [Prior art documents] [Patent documents]

[0004] [Patent Document 1] Patent Publication No. 2021-22231 Summary of the Invention [Problem to be solved by the invention]

[0005] However, in order to create a web application, various specialized knowledge and technologies related to the web (such as server technology, client technology, etc.) are required. Therefore, for AI developers who are not web experts, creating a web application for an AI system is difficult and a burdensome task.

[0006] In view of the above-mentioned problems, an object of the present invention is to provide an information processing device, an information processing method, and a program that can easily turn a system using a trained model into a web application. [Means for solving the problem]

[0007] In order to solve the above-mentioned problems, an information processing device according to one embodiment of the present invention is an information processing device that uses input information input by a user to a web-based user interface as input to a trained model and executes processing using the trained model, and is equipped with a definition information acquisition unit that acquires definition information that defines the input information to be input to the trained model and output information to be output from the trained model, and a user interface generation unit that generates the user interface based on the acquired definition information, the user interface including at least an input area where the user inputs the input information and an output area where the output information is output.

[0008] An information processing method according to one embodiment of the present invention is an information processing method in an information processing device that uses input information input by a user to a web-based user interface as input to a trained model and executes processing using the trained model, and includes a step in which a definition information acquisition unit acquires definition information that defines the input information to be input to the trained model and output information to be output from the trained model, and a step in which a user interface generation unit generates the user interface based on the acquired definition information, the user interface including at least an input area into which the user inputs the input information and an output area into which the output information is output.

[0009] A program according to one embodiment of the present invention is a program for causing a computer to function as an information processing device that uses input information input by a user to a web-based user interface as input to a trained model and executes processing using the trained model, and causes the computer to function as: a definition information acquisition step for acquiring definition information that defines the input information to be input to the trained model and output information to be output from the trained model; and a user interface generation step for generating the user interface based on the acquired definition information, the user interface including at least an input area where the user inputs the input information and an output area where the output information is output. Effect of the Invention

[0010] According to the present invention, a system using a trained model can be easily turned into a web application. [Brief description of the drawings]

[0011] [Figure 1] 1 is a block diagram showing an example of a configuration of an information processing system according to an embodiment of the present invention. [Diagram 2] FIG. 4 is a diagram showing an example of a display screen according to an embodiment of the present invention. [Diagram 3] FIG. 11 is a sequence diagram showing an example of a flow of a registration process according to an embodiment of the present invention. [Figure 4] 11 is a flowchart showing an example of the flow of a user interface generation process according to the embodiment of the present invention. [Diagram 5] FIG. 11 is a sequence diagram showing an example of a flow of an execution process according to the embodiment of the present invention. DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS

[0012] Hereinafter, embodiments of the present invention will be described in detail with reference to the drawings. The present invention relates to a system using AI (Artificial Intelligence) (hereinafter, also referred to as an "AI system"). Specifically, in this embodiment, an information processing system (an example of an AI system) using a trained model (an example of an AI model) will be described.

[0013] <<1. Information Processing System Configuration>> The configuration of an information processing system according to an embodiment of the present invention will be described with reference to Fig. 1. Fig. 1 is a block diagram showing an example of the configuration of an information processing system according to an embodiment of the present invention. As shown in Fig. 1, an information processing system 1 according to this embodiment includes a developer terminal 10, a user terminal 20, and a server device 30 (an example of an information processing device).

[0014] <1. Developer Terminal 10> The developer terminal 10 is a terminal used by a developer for various developments. The developer terminal 10 according to this embodiment is, for example, a PC (Personal Computer). Note that the developer terminal 10 is not limited to a PC, and may be a smartphone, a tablet terminal, or the like. The developer terminal 10 is communicably connected to the server device 30 via the network NW.

[0015] A developer, for example, uses a developer terminal 10 to develop a trained model and a program (hereinafter also referred to as an "execution program") that executes processing using the trained model. In the following, in this embodiment, an example will be described in which the trained model is a model for determining the possibility of blockage in pouring lightweight concrete pipes. Also, an example will be described in which the execution program is a program that executes PoC (Proof of Concept) for the trained model. Note that the execution program is a program for executing inference using the trained model, and is not limited to an example used for executing PoC. For example, the execution program may be used in the actual operation of the trained model.

[0016] A developer accesses a system for executing a developed executable program (hereinafter also referred to as an "execution environment system") via a Web browser. When the developer accesses the execution environment system, a Web-based user interface is displayed on the developer terminal 10. Specifically, a start-up screen of the execution environment system is displayed. The developer operates the displayed user interface to display a registration screen and performs registration processing for each trained model. In the registration processing, the developer registers a project, for example, for each trained model. In registering a project, the developer inputs definition information and uploads the trained model and execution program that he or she has developed.

[0017] The definition information is information that defines a project. The definition information mainly defines input information to be input to a trained model and output information to be output from the trained model. Specifically, the definition information includes a project code, a project name, a project description, an execution program name, a model name, the number of input parameters, an input parameter name list, the number of output parameters, and an output parameter name list.

[0018] The project code is identification information for uniquely identifying each project. As an example, the project name is "A003". In this embodiment, the project code is information assigned by the system. Therefore, the developer inputs eight items excluding the project code out of the nine items mentioned above. The project name indicates the name of the project. For example, the project name is "concreat". The project description indicates a sentence that describes the content of the project. For example, the project description is "Determine the possibility of clogging in lightweight concrete pipe casting." The execution program name indicates the name of the execution program to be executed. As an example, the execution program name is "concreat.py". The model name indicates the name of the trained model to be executed. For example, the model name is "concreat.pkl". The number of input parameters indicates the number of pieces of input information to be input to the trained model. As an example, the number of input parameters is "20". The input parameter name list indicates the names of the input parameters. For example, the input parameter name list may include "casting month, cement volume, slump, air volume, air volume aggregate correction coefficient, unit water volume, water absorption rate of coarse aggregate, water absorption rate of fine aggregate," etc. The number of output parameters indicates the number of pieces of output information output from the trained model. As an example, the number of output parameters is "1". The output parameter name list indicates the names of the output parameters. As an example, the output parameter name list indicates "blocking probability."

[0019] <2. User terminal 20> The user terminal 20 is a terminal used by a user (an example of a user in the claims) to use the runtime environment system. The user terminal 20 according to the present embodiment is, for example, a smartphone. Note that the user terminal 20 is not limited to a smartphone, and may be a tablet terminal, a PC, or the like. The user terminal 20 is communicably connected to the server device 30 via the network NW.

[0020] A user can access the runtime environment system via a Web browser. When the user accesses the runtime environment system, a Web-based user interface is displayed on the user terminal 20. Specifically, a start-up screen of the runtime environment system is displayed. The user operates the displayed user interface to display the execution screen and executes the trained model. In the execution process, the user selects the project to be executed from among registered projects, for example. After selecting the project, the user inputs input information to be input to the trained model and performs execution operations. After execution, output information indicating the execution results is output to the user terminal 20. The output information is displayed, for example, on a display unit (not shown) of the user terminal 20.

[0021] The developer terminal 10 and the user terminal 20 may be the same terminal or different terminals, regardless of whether the developer and the user are the same person. In the following, in this embodiment, if the developer and the user are the same person, the developer terminal 10 and the user terminal 20 are the same terminal, and if the developer and the user are different people, the developer terminal 10 and the user terminal 20 are different terminals.

[0022] <3. Server device 30> The server device 30 is a server device that executes various processes. The server device 30 according to the present embodiment is realized by, for example, a PC. The server device 30 is communicably connected to the developer terminal 10 and the user terminal 20 via the network NW.

[0023] The server device 30 is a device that uses input information entered by a user into a web-based user interface as input to a trained model and executes processing using the trained model. As shown in FIG. 1, the server device 30 includes a communication unit 31, a storage unit 32, and a control unit 33.

[0024] (1) Communications Department 31 The communication unit 31 has a function of transmitting and receiving various information. For example, the communication unit 31 transmits and receives information to and from the developer terminal 10 and the user terminal 20 via the network NW.

[0025] (2) Storage section 32 The storage unit 32 has a function of storing various information. The storage unit 32 is configured by a storage medium provided in the server device 30 as hardware, such as a hard disk drive (HDD), a solid state drive (SSD), a flash memory, an electrically erasable programmable read only memory (EEPROM), a random access read / write memory (RAM), a read only memory (ROM), or any combination of these storage media. 1, the storage unit 32 stores various DBs (databases). Specifically, the storage unit 32 stores a definition information DB 321, a trained model DB 322, an execution program DB 323, and a code DB 324.

[0026] The definition information DB 321 is a database that stores definition information. The trained model DB 322 is a database that stores trained models. For example, the trained model DB 322 stores trained models uploaded by developers in association with definition information of the trained models. The execution program DB 323 is a database that stores execution programs. For example, the execution program DB 323 stores an execution program uploaded by a developer in association with a trained model that is executed by the execution program. The code DB 324 is a database that stores codes of user interfaces. For example, the code DB 324 stores codes of execution screens (user interfaces) generated for each trained model in association with the trained models.

[0027] The trained model DB322 may be configured to store only the file names of the trained models. In this case, the files of the trained models themselves are stored (memorized) in a folder separately prepared in the storage unit 32. Similarly, the execution program DB323 may be configured to store only the file names of the execution programs. In this case, the files of the execution programs themselves are stored (memorized) in a folder separately prepared in the storage unit 32. The folders in which the trained models and execution programs are stored are folders for storing information on each project for each project indicated by the definition information (hereinafter also referred to as "project folders").

[0028] (3) The control unit 33 The control unit 33 has a function of controlling the overall operation of the server device 30. The control unit 33 is realized, for example, by causing a CPU (Central Processing Unit) provided as hardware in the server device 30 to execute a program. As shown in FIG. 1, the control unit 33 includes a definition information acquisition unit 331 , an input information acquisition unit 332 , a registration unit 333 , a folder processing unit 334 , a user interface generation unit 335 , a program execution unit 336 , and a display processing unit 337 .

[0029] (3-1) Definition information acquisition unit 331 The definition information acquisition unit 331 has a function of acquiring definition information. For example, the definition information acquisition unit 331 acquires definition information input by a developer into a definition information input field on a registration screen. Specifically, the definition information acquisition unit 331 acquires definition information corresponding to a trained model uploaded by a developer from the definition information DB 321 when a user interface generation unit 335 described later generates a user interface.

[0030] (3-2) Input information acquisition unit 332 The input information acquisition unit 332 has a function of acquiring input information. For example, the input information acquisition unit 332 acquires input information entered by a user in an input information input field (input area) of an execution screen (user interface). Specifically, the input information acquisition unit 332 acquires the input information entered by the user when the program execution unit 336 described later executes an execution program.

[0031] (3-3) Registration Section 333 The registration unit 333 has a function of registering various types of information. For example, the registration unit 333 registers definition information input by a developer, and a trained model and an execution program uploaded by a developer.

[0032] (3-4) Folder processing unit 334 The folder processing unit 334 has a function of performing processing related to folders. For example, the folder processing unit 334 creates a project folder for each project indicated by the registered definition information. The folder processing unit 334 also creates a folder (template folder) for storing a user interface generated by the user interface generating unit 335 described later as a template. The folder processing unit 334 may control the input and output of information to and from each created folder.

[0033] (3-5) User interface generation unit 335 The user interface generating unit 335 has a function of generating a user interface. Based on the acquired definition information, the user interface generating unit 335 generates a user interface including at least an input area where the user inputs input information and an output area where output information is output. The user interface generating unit 335 generates the user interface using, for example, HTML (Hyper Text Markup Language) or a script. The user interface generating unit 335 may generate source code or script code. In this way, the user interface is automatically and dynamically generated by the user interface generating unit 335.

[0034] (3-6) Program Execution Unit 336 The program execution unit 336 has a function of executing an execution program. For example, the program execution unit 336 executes an execution program corresponding to the input information based on acquired input information. For example, the program execution unit 336 executes an execution program corresponding to the input information based on a file name searched from the execution program DB 323.

[0035] (3-7) Display processing unit 337 The display processing unit 337 has a function of controlling the display on each terminal. For example, the display processing unit 337 causes the developer terminal 10 and the user terminal 20 to display a start-up screen. The display processing unit 337 also causes the developer terminal 10 to display a registration screen. The display processing unit 337 also causes the user terminal 20 to display an execution screen.

[0036] Here, the display screen according to the present embodiment will be described with reference to Fig. 2. Fig. 2 is a diagram showing an example of the display screen according to the embodiment of the present invention. Fig. 2 shows a start-up screen 40, a registration screen 50, and an execution screen 60 as examples of the display screen.

[0037] The start-up screen 40 is an example of a screen that is displayed when the execution environment system is accessed. As shown in Fig. 2, the start-up screen 40 displays, as an example, a project registration button 41, a database management button 42, and a project list 43. The project registration button 41 is a button that transitions the screen from the start-up screen 40 to a registration screen 50 when pressed. The database management button 42 is a button that, when pressed, transitions the screen from the start-up screen 40 to a database management screen (not shown). The project list 43 shows the names of registered projects. When a displayed project name is selected, the screen transitions to the corresponding execution screen 60. For example, as shown in FIG. 2, when "(3) Project III" is selected, the execution screen 60 of Project III is displayed.

[0038] The registration screen 50 is a screen for registering definition information for each project. As shown in Fig. 2, as an example, the registration screen 50 displays definition information input fields 51-1 to 51-N (N is a natural number) and a registration button 52. When a developer inputs definition information into the definition information input fields 51-1 to 51-N and presses the registration button 52, the input definition information is registered.

[0039] The execution screen 60 is a screen for executing a selected project (i.e., executing an execution program). As shown in Fig. 2, the execution screen 60 displays, as an example, input information input fields 61-1 to 61-M (M is a natural number), an execute button 62, and an execution result output field 63. When a user inputs input information in the input information input fields 61-1 to 61-M (input area) and presses the execute button 62, the execution result is displayed in the execution result output field 63 (output area). As described above, since the output parameter name in this embodiment is "blocking probability," the execution result output field 63 indicates that the blocking probability is "x%."

[0040] <4. Processing flow> The flow of processing according to the embodiment of the present invention will be described with reference to Fig. 3 to Fig. 5. Below, the registration processing, user interface generation processing, and execution processing will be described with reference to each figure.

[0041] (1) Registration process The flow of the registration process will be described with reference to Fig. 3. Fig. 3 is a sequence diagram showing an example of the flow of the registration process according to the embodiment of the present invention. As shown in FIG. 3, first, the developer develops a trained model and an execution program for the trained model in advance (steps S101 and S102). The developer operates the developer terminal 10 to perform an operation to access the runtime environment system (step S103). When the developer terminal 10 accepts an input of an operation to access the runtime environment system by the developer, the developer terminal 10 transmits a request to access the runtime environment system to the server device 30 (step S104). It is assumed that the communication unit 31 of the server device 30 receives the request from the developer terminal 10. In this case, the display processing unit 337 of the server device 30 transmits information for displaying a start-up screen of the runtime environment system from the communication unit 31 to the developer terminal 10 in response to the request (step S105). The developer terminal 10 displays the start-up screen based on the information received from the server device 30 (step S106).

[0042] The developer performs an operation on the start-up screen to access the registration screen (step S111). When the developer terminal 10 accepts the input of the operation to access the registration screen by the developer, it transmits a request to the server device 30 to access the registration screen (step S112). It is assumed that the communication unit 31 of the server device 30 receives the request from the developer terminal 10. In this case, the display processing unit 337 of the server device 30 transmits information for displaying the registration screen from the communication unit 31 to the developer terminal 10 as a response to the request (step S113). The developer terminal 10 displays the registration screen based on the information received from the server device 30 (step S114). The developer inputs definition information to the registration screen (step S115). When the developer terminal 10 accepts the input of the definition information by the developer, it transmits the definition information to the server device 30 (step S116). When the communication unit 31 receives the definition information from the developer terminal 10, the registration unit 333 of the server device 30 writes and stores the definition information received by the communication unit 31 in the definition information DB 321 of the storage unit 32 (step S117). The folder processing unit 334 of the server device 30 creates a project folder for the project in which the definition information registered by the developer has been registered (step S118).

[0043] The developer uploads the trained model and the execution program on the registration screen (step S121). When the developer terminal 10 accepts the upload operation by the developer, it transmits the trained model and the execution program to the server device 30 (step S122). When the communication unit 31 receives the trained model and the execution program from the developer terminal 10, the registration unit 333 writes and stores the trained model and the execution program received by the communication unit 31 in the trained model DB 322 and the execution program DB 323 of the storage unit 32, respectively. In addition, the registration unit 333 stores the trained model and the execution program in a project folder (step S123). After storing, the user interface generation unit 335 of the server device 30 executes a user interface generation process (step S124). Details of the user interface generation process will be described later. The user interface generation unit 335 writes and stores the code of the generated user interface in the code DB 324 of the storage unit 32. Furthermore, the user interface generating unit 335 stores the code of the generated user interface in the template folder (step S125).

[0044] (2) User interface generation process FIG. 4 is a flowchart showing an example of the flow of a user interface generation process according to an embodiment of the present invention. As shown in FIG. 4, first, the definition information acquisition unit 331 acquires definition information from the definition information DB 321 (step S201). Next, the user interface generation unit 335 adds the UI code of HTML to create a new file (step S202). Next, the user interface generation unit 335 generates a fixed code such as a UI library definition (step S203). Next, the user interface generation unit 335 generates a variable name receiving code for receiving the variable name of the input information (step S204) and an input code for displaying an input area for the input information based on the definition information (step S205). The user interface generation unit 335 repeats the generation of the variable name receiving code and the input code the number of times indicated by the number of input parameters. Next, the user interface generation unit 335 generates a cancellation code (step S206). Next, the user interface generation unit 335 generates an output result display code for displaying an output area of ​​the execution result based on the definition information (step S207). Next, the user interface generating unit 335 generates a code for reading the library (step S208). The fixed code is written, for example, in the HEAD tag of HTML. The variable name receiving code, input code, cancellation code, output result display code, and library reading code are written, for example, in the BODY tag of HTML.

[0045] (3) Execution process FIG. 5 is a sequence diagram showing an example of the flow of execution processing according to the embodiment of the present invention. As shown in FIG. 5, the user operates the user terminal 20 to access the runtime environment system (step S301). When the user terminal 20 accepts the user's input for the operation to access the runtime environment system, it transmits a request to access the runtime environment system to the server device 30 (step S302). It is assumed that the communication unit 31 receives a request from the developer terminal 10. In this case, the display processing unit 337 transmits information for displaying a start-up screen of the runtime environment system from the communication unit 31 to the developer terminal 10 in response to the request (step S303). The user terminal 20 displays the start-up screen based on the information received from the server device 30 (step S304).

[0046] The user performs an operation on the start-up screen to access the execution screen (step S311). When the user terminal 20 accepts the user's input of an operation to access the execution screen, it transmits a request to the server device 30 to access the execution screen (step S312). It is assumed that the communication unit 31 receives the request from the user terminal 20. In this case, the display processing unit 337 transmits information for displaying the execution screen from the communication unit 31 to the user terminal 20 in response to the request (step S313). The user terminal 20 displays the execution screen based on the information received from the server device 30 (step S314).

[0047] The user inputs input information on the execution screen and performs an execution operation (step S321). When the user terminal 20 accepts the execution operation by the user, it transmits the input information and an execution request to the server device 30 (step S322). When the communication unit 31 receives the execution request from the user terminal 20, the input information acquisition unit 332 of the server device 30 acquires the input information from the communication unit 31. The program execution unit 336 of the server device 30 executes the execution program of the project corresponding to the input information based on the input information (step S323). After the execution, the display processing unit 337 transmits the execution result from the communication unit 31 to the user terminal 20 (step S324). The user terminal 20 displays the execution result received from the server device 30 (step S325).

[0048] As described above, the server device 30 according to the present embodiment is an information processing device that uses input information input by a user to a Web-based user interface as an input to a trained model and executes processing using the trained model. As described above, the server device 30 includes the definition information acquisition unit 331 and the user interface generation unit 335. The definition information acquisition unit 331 acquires definition information that defines input information to be input to the trained model and output information to be output from the trained model. The user interface generating unit 335 generates a user interface based on the acquired definition information, the user interface including at least an input area where the user inputs input information and an output area where output information is output.

[0049] With this configuration, the server device 30 generates a web-based user interface according to the input / output items of the trained model to be used. This eliminates the need for users such as developers and users to manually code according to the input / output items of the trained model to be used when generating a web-based user interface. Therefore, even an AI developer who is not a web expert can easily create a web application of an AI system.

[0050] As described above, the server device 30 according to this embodiment makes it possible to easily turn a system using a trained model into a web application. In addition, by reducing the amount of manual coding, the time required to turn an AI system into a web application can be shortened, making it possible to utilize the AI ​​system more quickly. It also makes it possible to reduce the number of people who do manual coding and shorten the development time for the entire AI system, thereby reducing the cost of developing an AI system.

[0051] The embodiment of the present invention has been described above. Note that a part or the whole of the server device 30 in the above-mentioned embodiment may be realized by a computer. In that case, a program for realizing this function may be recorded in a computer-readable recording medium, and the program recorded in the recording medium may be read into a computer system and executed to realize the function. Note that the "computer system" here includes hardware such as an OS and peripheral devices. In addition, the "computer-readable recording medium" refers to a portable medium such as a flexible disk, an optical magnetic disk, a ROM, a CD-ROM, and a storage device such as a hard disk built into a computer system. Furthermore, the "computer-readable recording medium" may include a device that dynamically holds a program for a short period of time, such as a communication line when a program is transmitted via a network such as the Internet or a communication line such as a telephone line, and a device that holds a program for a certain period of time, such as a volatile memory inside a computer system that is a server or client in that case. In addition, the above-mentioned program may be a program for realizing a part of the above-mentioned function, or may be a program that can realize the above-mentioned function in combination with a program already recorded in the computer system, or may be a program that is realized using a programmable logic device such as an FPGA (Field Programmable Gate Array).

[0052] Although the embodiment of the present invention has been described in detail above with reference to the drawings, the specific configuration is not limited to the above, and various design changes, etc. are possible within the scope that does not deviate from the gist of the present invention. [Explanation of symbols]

[0053] 1...information processing system, 10...developer terminal, 20...user terminal, 30...server device, 31...communication unit, 32...storage unit, 33...control unit, 40...startup screen, 50...registration screen, 60...execution screen, 321...definition information DB, 322...trained model DB, 323...execution program DB, 324...code DB, 331...definition information acquisition unit, 332...input information acquisition unit, 333...registration unit, 334...folder processing unit, 335...user interface generation unit, 336...program execution unit, 337...display processing unit, NW...network

Claims

1. An information processing device that uses input information input by a user to a web-based user interface as an input to a trained model and executes a process using the trained model, A definition information acquisition unit that acquires definition information that defines the input information to be input to the trained model and the output information to be output from the trained model; a user interface generating unit that generates the user interface based on the acquired definition information, the user interface including at least an input area where the user inputs the input information and an output area where the output information is output; An information processing device comprising:

2. A storage unit that stores the trained model and an execution program for executing a process using the trained model; an input information acquisition unit that acquires the input information input into the input area of ​​the user interface; a program execution unit that executes the execution program corresponding to the acquired input information based on the acquired input information; The information processing device according to claim 1 , further comprising:

3. The execution program executes a Proof of Concept (PoC) for the trained model. The information processing device according to claim 2 .

4. The user interface generation unit generates the user interface using HTML (Hyper Text Markup Language) or a script. The information processing device according to claim 1 .

5. An information processing method in an information processing device that uses input information input by a user to a web-based user interface as an input to a trained model and executes processing using the trained model, A step in which a definition information acquisition unit acquires definition information that defines the input information to be input to the trained model and the output information to be output from the trained model; a step of generating a user interface by a user interface generating unit based on the acquired definition information, the user interface including at least an input area where the user inputs the input information and an output area where the output information is output; An information processing method comprising:

6. A program for causing a computer to function as an information processing device that uses input information input by a user to a web-based user interface as an input to a trained model and executes processing using the trained model, The computer, A definition information acquisition step of acquiring definition information that defines the input information to be input to the trained model and the output information to be output from the trained model; a user interface generating step of generating the user interface based on the acquired definition information, the user interface including at least an input area where the user inputs the input information and an output area where the output information is output; A program to function as a

Images