System, method, and program

The system facilitates the creation of applications using generative AI by allowing users to define structures and processing transitions, enhancing AI interpretation and reducing errors through predefined candidates and explanatory information.

JP7799134B1Active Publication Date: 2026-01-16유겐가이샤티아이에스
View PDF 5 Cites 0 Cited by

Patent Information

Application Number
JP2025163516
Authority / Receiving Office
JP · JP
Patent Type
Patents
Current Assignee / Owner
Filing Date
2025-09-30
Publication Date
2026-01-16
Estimated Expiration
2045-09-30

AI Technical Summary

Technical Problem

Existing systems that allow users to create applications using generative AI without coding struggle to accurately reflect user requests and require complex operations.

Method used

A system that includes a structural information acquisition unit, processing designation information, and transition information acquisition unit to allow users to define application structures and processing transitions using predefined processing candidates, with explanatory information to enhance AI interpretation.

Benefits of technology

Enables the creation of applications using generative AI with simple operations that accurately reflect user needs, reducing application errors and improving user understanding of the processing.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure 0007799134000001_ABST
    Figure 0007799134000001_ABST
Patent Text Reader

Abstract

A system, method, and program are provided that enable applications using generative AI to be created flexibly with simple operations. [Solution] An application management device 101 for creating an application includes a structural information acquisition unit 10142 that acquires structural information defining the structure of multiple variable data to be processed in the application, processing specification information that is assigned to each of multiple processing blocks and specifies a process selected from multiple pre-defined processing candidates including processing using a generation AI, and data specification information that specifies the variable data to be processed by each of the multiple processing blocks as a variable having a structure defined by the structural information, and is equipped with a processing information acquisition unit 10143 that acquires processing information defining the processing in the application, and a transition information acquisition unit 10144 that acquires transition information that defines the processing transition between multiple processing blocks.
Need to check novelty before this filing date? Find Prior Art

Description

[Technical Field]

[0001] The present invention relates to a system, a method, and a program. [Background technology]

[0002] In recent years, there has been a system that allows users to create applications without writing source code, as one of the systems that create applications using generative AI. Patent Document 1 describes a system that allows users without programming skills to generate application creation script code for business automation. [Prior art documents] [Patent documents]

[0003] [Patent Document 1] Japanese Patent Application Laid-Open No. 2025-48865 Summary of the Invention [Problem to be solved by the invention]

[0004] By using a system that allows users to create applications using generative AI without coding, even users without specialized programming skills can easily create applications. However, it has been difficult to create applications using generative AI that can be created without coding while reflecting user requests.

[0005] Therefore, an object of the present invention is to provide a system, method, and program that allows applications using generative AI to be created with simple operations while reflecting user requests. [Means for solving the problem]

[0006] A system for creating an application according to one embodiment of the present invention comprises a structural information acquisition unit that acquires structural information defining the structure of multiple variable data to be processed in the application, processing designation information that is assigned to each of multiple processing blocks and designates a process selected from multiple pre-defined processing candidates including a process using a generation AI, and data designation information that designates the variable data to be processed by each of the multiple processing blocks as a variable having a structure defined by the structural information, and a transition information acquisition unit that acquires transition information that defines the processing in the application, and the transition information that defines the processing transitions between the multiple processing blocks.

[0007] According to this aspect, a user can set the processing in the application by simply selecting from a predefined set of processing candidates. Furthermore, since the user can define the structure of the data processed in the application, it is possible to define the data structure and assign variable data to processing blocks in accordance with the user's needs. This allows an application using a generative AI to be created with simple operations, reflecting the user's needs.

[0008] In the above aspect, the structural information may include explanatory information that indicates an explanation of the structure, and the explanatory information may be included in the input to the generation AI in the process using the generation AI.

[0009] According to this aspect, the prompt input to the generation AI includes an explanation of the structure, allowing the generation AI to accurately interpret the prompt, thereby improving the accuracy of the output from the generation AI.

[0010] In the above aspect, the system may include an explanation display unit that displays explanation information to the user.

[0011] According to this aspect, the user can grasp the meaning of the structural information from the explanatory information, and thus can understand the output results from the application and the processing executed by the application.

[0012] In the above aspect, the processing using the generation AI may input a prompt to the generation AI requesting that the generation AI output structural correspondence information corresponding to the structure of a variable specified as an output for the AI ​​processing block to which the processing using the generation AI is assigned.

[0013] According to this aspect, the prompts can be interpreted accurately by the generated AI, and the output from the generated AI can be obtained in a form that meets the processing objectives of the AI ​​processing block.

[0014] In the above aspect, the processing in the AI ​​processing block may include re-inputting a prompt to the generation AI if the structure correspondence information is not output from the generation AI, and if the structure correspondence information is not output from the generation AI after a predetermined number of re-inputs, the variable designated as the output may be updated based on the information output from the generation AI.

[0015] According to this aspect, even if structural correspondence information corresponding to the structure of a variable specified as output is not output, application errors due to the inability to update the variable can be avoided, and the variable can be updated using the provisionally output information, thereby completing the processing.

[0016] In the above aspect, the structure information may include first structure information indicating a structure of variable data preset in the application and second structure information indicating a structure selected from a plurality of candidate variable data structures.

[0017] According to this aspect, among the structural information in an application, structural information that is commonly or frequently used in each processing block is prepared in advance as first structural information, and structural information according to the purpose of each application can be set by the user as second structural information, thereby reducing the burden on the user in setting structural information and enabling the user to set structural information that suits their purpose.

[0018] A method for creating an application according to one embodiment of the present invention includes a computer acquiring structural information that defines the structure of multiple variable data items to be processed in the application, processing designation information that is assigned to each of multiple processing blocks and that specifies a process selected from multiple pre-defined processing candidates including a process using a generation AI, and data designation information that designates the variable data items to be processed by each of the multiple processing blocks as variables having a structure defined by the structural information, and acquiring transition information that defines the processing in the application.

[0019] A program for creating an application according to one embodiment of the present invention includes, on a computer, acquiring structural information that defines the structure of multiple variable data items to be processed in the application, processing designation information that is assigned to each of multiple processing blocks and that designates a process selected from multiple pre-defined processing candidates including a process using a generation AI, and data designation information that designates the variable data to be processed by each of the multiple processing blocks as a variable having a structure defined by the structural information, and acquiring processing information that defines the processing in the application, and acquiring transition information that defines the processing transitions between the multiple processing blocks. [Effects of the Invention]

[0020] According to the present invention, it is possible to provide a system, method, and program that can create applications using generation AI with simple operations, reflecting user requests. [Brief explanation of the drawings]

[0021] [Figure 1] FIG. 1 is a diagram illustrating an overview of an application creation system. [Figure 2] FIG. 2 is a block diagram of each device related to the application creation system. [Figure 3] FIG. 2 is a diagram illustrating an outline of processing in the application creation system. [Figure 4] FIG. 10 is a diagram showing an example of a completed application screen displayed on a user terminal. [Figure 5] FIG. 10 is a diagram illustrating an example of an application creation screen displayed on a user terminal. [Figure 6] FIG. 10 is a diagram illustrating an example of an application creation screen displayed on a user terminal. [Figure 7] FIG. 10 is a diagram illustrating an example of an application creation screen displayed on a user terminal. DETAILED DESCRIPTION OF THE INVENTION

[0022] Hereinafter, an embodiment of the system of the present invention will be described in detail with reference to the drawings. In each drawing, components with the same reference numerals have the same or similar configurations.

[0023] <<Configuration Overview>> An overview of the application creation system 10 will be described with reference to Fig. 1. In the following description, the creator of an application will be referred to as a user, and the user of the application will be referred to as a user.

[0024] The application creation system 10 includes an application management device 101, a generation device 201, and user terminals 301a to 301n. When no distinction is necessary, the user terminals 301a to 301n will be collectively referred to as user terminals 301.

[0025] The application management device 101 is communicably connected to the generation device 201 and the user terminal 301 via a network N.

[0026] The generation device 201 uses an AI analysis model to process an application based on a processing method set by a user. The AI ​​analysis model is a learning model that has undergone machine learning, and outputs output data corresponding to input information such as natural language or data, for example, as text, data, or voice. The AI ​​analysis model is, for example, a generative AI model such as a large-scale language model. The type of AI analysis model used in an application may be selected by the user or may be selected by the application management device 101.

[0027] The user terminal 301 is an information processing terminal operated by a user, and is, for example, a smartphone, a tablet terminal, or a personal computer (for example, a desktop or laptop).

[0028] Referring to FIG. 2, the application management device 101, the generation device 201, and the user terminal 301 in the application creation system 10 will be described.

[0029] The application management device 101 includes a communication unit 1011 , a storage unit 1012 , an operation information acquisition unit 1013 , an application creation unit 1014 , a processing information transfer unit 1015 , and a generated data reception unit 1016 .

[0030] The communication unit 1011 controls communication between the application management device 101 and the generation device 201 and the user terminal 301 .

[0031] The storage unit 1012 stores various types of information used in the processing of the application management apparatus 101 .

[0032] The operation information acquisition unit 1013 acquires various information based on operation input at the user terminal 301 .

[0033] The application creation unit 1014 has a first basic information acquisition unit 10141, a structure information acquisition unit 10142, a processing information acquisition unit 10143, a transition information acquisition unit 10144, and a second basic information acquisition unit 10145. Each acquisition unit acquires information input to the user terminal 301. In addition, the application creation unit 1014 refers to definition items stored in the storage unit 1012 that are necessary when creating an application.

[0034] The first basic information acquisition unit 10141 acquires first basic information including, for example, the name of the application, a description of the application, display elements, accessible users, and setting information of the application icon. The first basic information acquisition unit 10141 stores the acquired first basic information in the storage unit 1012 in association with information identifying the application.

[0035] The structure information acquisition unit 10142 acquires information about the structure set by the user from among multiple pieces of variable data processed by the application. The structure information includes at least a structure ID and a structure type.

[0036] The structure information includes first structure information indicating a variable data structure preset in the application and second structure information indicating a structure selected from multiple variable data structure candidates. The first structure information is structure information that is common to all applications. For example, the structure ID and structure type of the first structure information are predetermined. For example, the first structure information has a structure ID of "query" and a structure type of "str" ​​(character string), and is string-type structure information that stores the user's initial input to the application. Because the structure ID and structure type of the first structure information are defined in advance, the user can omit the operation of setting the structure ID and structure type when specifying the first structure information. The second structure information is structure information that is defined by the user setting the structure ID and structure type. The setting process of the second structure information will be described later.

[0037] The structural information may also include explanatory text that explains the structural information. The explanatory text is used to display the explanatory text to users who use the application. The explanatory text can also be used as an input prompt for the AI ​​analysis model.

[0038] The first structure information is stored in advance in the storage unit 1012. The structure information acquisition unit 10142 stores the second structure information set by the user in the storage unit 1012 in association with information for identifying the application.

[0039] The processing information acquisition unit 10143 acquires processing information that defines processing in an application, and includes processing specification information that is assigned to each of a plurality of processing blocks and specifies a processing selected from a plurality of pre-defined processing candidates including processing using a generation AI, and data specification information that specifies variable data to be processed by each of the plurality of processing blocks as variables having a structure defined by structural information.

[0040] A processing block is a unit in which processing is executed in an application. An application operates by multiple processing blocks working together to execute their respective processes. A processing block is defined based on process specification information and data specification information.

[0041] The process specification information is information that specifies a process that uses a generation AI, such as a "multiple RAG action" or an "LLM action." The process specified by the process specification information may also include a process that does not use a generation AI.

[0042] "Multiple RAG action" is a process of selecting and searching appropriate data from multiple configured RAG (Retrieval-Augmented Generation) datasets and web information in response to a query input by a user, and outputting a response from the analytical model unit 2013.

[0043] An "LLM action" is a process that uses the analytical model unit 2013 to obtain output based on a specified prompt and update the value of a variable.

[0044] The data specification information specifies the variables to be processed in each processing block. The variables have a structure specified by the structure information. For example, if a "multiple RAG action" or "LLM action" is specified as a process, a variable to store the answer is specified.

[0045] Processing using the analytical model unit 2013 includes inputting a prompt to the analytical model unit 2013 requesting that the analytical model unit 2013 output structure correspondence information corresponding to the structure of a variable specified as output for a processing block to which processing using the analytical model unit 2013 is assigned.

[0046] Specifically, the prompt input to the analysis model unit 2013 requests an output corresponding to a specified variable as structure-corresponding information. For example, assume that structural information called "search results" has elements such as "search perspective," "search results," and "result likelihood," and is specified as a variable. In this case, the prompt requests that information capable of updating the "search perspective," "search results," and "result likelihood" be output as structure-corresponding information. For example, a prompt such as "Please output the search processing execution results so that they include the search perspective, search results, and result likelihood" is input. The output is obtained, for example, as "Search perspective: Search from document data where files exist, Search results: (searched document name), Result likelihood: Sufficiently meets the user's requirements," and the information on the "search perspective," "search results," and "result likelihood" of the structural information is updated, respectively.

[0047] The processing information acquisition unit 10143 presents a plurality of processing specification information candidates stored in the storage unit 1012 to the user, and acquires the processing selected by the user as processing specification information. The processing information acquisition unit 10143 acquires a structure ID that specifies the structure information stored in the storage unit 1012 as data specification information.

[0048] By associating data specification information with processing specification information, for example, in processing involving analysis results by a generation AI, the structure of the data output can be set based on the data specification information. This allows application users to specify output that is easy to operate, leading to improved work efficiency for users.

[0049] In the above example, the structure correspondence information is appropriately acquired. However, it is also possible that the output of the analytical model unit 2013 is not suitable for the structure correspondence information. In this case, the processing using the analytical model unit 2013 may include re-inputting a prompt to the analytical model unit 2013 when the structure correspondence information is not output from the analytical model unit 2013, and may also include updating the variables specified as output based on the information output from the analytical model unit 2013 when the structure correspondence information is not output from the analytical model unit 2013 after a predetermined number of re-inputs. This processing can be performed by setting whether or not repeated input is permitted and the maximum number of repeated inputs in the information that sets the processing specification information. This allows the structural information to be updated based on the answer generated in the last processing when the predetermined number of processing times is exceeded, thereby preventing the structural information from being unable to be updated due to an output error from the analytical model unit 2013, for example.

[0050] The transition information acquisition unit 10144 acquires transition information that defines the transition of processing between a plurality of processing blocks.

[0051] The transition information includes information for selecting a processing block from which the transition is to occur, information for selecting a processing block to which the transition is to occur, and a transition condition. For example, when processing by a processing block is normally performed and target structural information is updated, information for setting a transition to the next processing block is acquired as the transition information.

[0052] The second basic information acquisition unit 10145 acquires second basic information including, for example, the structure of variable data used when starting an application and identifier information for authenticating or controlling access.

[0053] The processing information transfer unit 1015 acquires information used for processing by the generation device 201 from the user's terminal when using the application, including operation information input from the user's terminal and input content and data used in the application, and transfers it to the generation device 201.

[0054] The generated data receiving unit 1016 receives the data generated by the generating device 201 .

[0055] The generating device 201 includes a communication unit 2011 , a processing information acquisition unit 2012 , an analysis model unit 2013 , and a generated data transmission unit 2014 .

[0056] The communication unit 2011 controls communication between the generation device 201 and the application management device 101 .

[0057] The processing information acquisition unit 2012 acquires, from the application management device 101, information used in processing by the generation device 201, including operation information input from the user's terminal and input contents and data used in the application.

[0058] The analysis model unit 2013 has an analysis unit 20131 and an answer generation unit 20132. The analysis unit 20131 processes the information acquired by the processing information acquisition unit 2012 and analyzes a prompt corresponding to structural information used as input. The answer generation unit 20132 generates an answer corresponding to structural information specified as output. The AI ​​analysis model used by the analysis model unit 2013 may be selected by the user if the user is allowed to select the model when creating the application.

[0059] The generated data transmission unit 2014 transmits the response data generated by the analysis model unit 2013 to the application management apparatus 101 .

[0060] The user terminal 301 includes a communication unit 3011 , an operation acquisition unit 3012 , a storage unit 3013 , and a display unit 3014 .

[0061] The communication unit 3011 controls communication between the user terminal 301 and the application management apparatus 101 .

[0062] The operation acquisition unit 3012 acquires an operation input from a user. The operation acquisition unit 3012 is, for example, a keyboard, a mouse, or a touch panel.

[0063] The storage unit 3013 stores various types of information used in processing by the user terminal 301 .

[0064] The display unit 3014 is a display of the user terminal 301 .

[0065] <<Processing Overview>> The process of creating an application in the application creation system 10 will be described with reference to Figures 3 to 7. As an example of an application to be created, a chat-style application as shown in application screen 401 of Figure 4 is shown. In the application shown in Figure 4, a "multiple RAG action" executes a process in which a response to an input from a user is output from multiple RAG data sets and web information.

[0066] In step S301, an operation to create a new application is performed on the user terminal 301.

[0067] In step S302, the user terminal 301 requests the application management apparatus 101 to create a new application.

[0068] In step S303, the first basic information acquisition unit 10141 creates a setting screen for the first basic information. Through the setting screen for the first basic information, the user can input setting items for information to be set as the first basic information. The setting items include, for example, items for setting the application name, application description, display elements, accessible users, and app icon. Information on the setting items for the first basic information is stored in the storage unit 1012.

[0069] In step S304, the first basic information acquisition unit 10141 transmits information on the setting screen for the first basic information to the user terminal 301.

[0070] In step S305, the user inputs the first basic information through the first basic information setting screen.

[0071] In step S 306 , the first basic information acquisition unit 10141 acquires the first basic information from the user terminal 301 .

[0072] In step S307, the first basic information acquisition unit 10141 stores the first basic information in association with the application.

[0073] In step S308, an operation is performed on the user terminal 301 to set the structure information of the variable data used in the operation of the application.

[0074] In step S309, the user terminal 301 requests the application management apparatus 101 to set the structure information.

[0075] In step S310, the structure information acquisition unit 10142 generates a structure information setting screen. The structure information setting screen includes structure information setting items and structure information type options. The structure information setting items are, for example, items for setting a structure ID, a structure information title, a structure information description, a structure type, an initial value assigned to the structure type, a flag for setting whether the contents of the structure information are displayed when the application is used, and a flag for setting whether the structure information is editable when the application is used. The structure type options are type candidates when the structure type is selected from predetermined data types (string type: str, numeric type: int, real number type: float, boolean type: bool, etc.). The structure type may include a structure type having multiple elements. Information on the structure information setting items and structure information type options is stored in the storage unit 1012.

[0076] In steps S311 and S312, the structure information acquisition unit 10142 transmits the setting items of the structure information and the options for the type of structure information to the user terminal 301. These processes are performed as a process of transmitting information on the structure information setting screen.

[0077] In step S313, the user inputs setting items and selects the type of structure based on the information acquired from the application management apparatus 101.

[0078] In steps S314 and S315, the user terminal 301 transmits the input setting items and the selected type of structure to the application management apparatus 101.

[0079] In step S316, the structure information acquisition unit 10142 stores the input information and the selected information as structure information.

[0080] 5 shows an example of a structure setting screen 501. The structure setting screen 501 has a common input area 5011, a type selection area 5012, and an input area by type 5013.

[0081] In the common input area 5011, for example, a structure ID, a title of the structure information, and a description of the structure information are input. In the example of FIG. 5, the structure ID is input as "webswitch," the title is input as "Web search," and the description is input as "Web search flag." In the type selection area 5012, for example, a structure type (structure format) is selected and input from options displayed in a pull-down format. In the example of FIG. 5, the type is input as "bool." In the type input area 5013, information according to the type of structure selected in the type selection area 5012 is input. For example, if the type is "bool," "false" is input as the initial value. In addition, in the type input area 5013, a flag that sets whether the contents of the structure information can be displayed when the application is used, and a flag that sets whether the structure information can be edited when the application is used are input, each indicating permission. Based on the structural information set in the example of Figure 5, a switch for "Web search" is displayed on the application screen 401 of Figure 4, and by turning the switch on or off, the user can set whether or not to take Web search results into consideration when answering a question.

[0082] In addition to the structural information shown in the example of Figure 5, other structural information is also acquired. For example, structural information of the structure in which the information referenced in the search is stored and structural information of the structure in which the answer to the question from the user is stored are respectively acquired. For example, structural information defining a structure with a structure ID of "rag_content", a title of "RAG content", a description of "Stores search results by RAG", and a type of "rag_context" is respectively acquired, and structural information defining a structure with a structure ID of "answer", a title of "answer", a description of "Answer based on search results", and a type of "str".

[0083] Returning to FIG. 3, in step S317, an operation is performed on the user terminal 301 to set the operation details of the application.

[0084] In step S318, the user terminal 301 requests the application management apparatus 101 to set the operation content.

[0085] In step S319, the processing information acquisition unit 10143 generates an operation content setting screen. The operation content setting screen includes operation content setting items and operation content type options. The operation content setting items are items for setting, for example, the ID of the processing block, the title of the processing block, the description of the processing block, the variable data to be updated that is assigned to the processing block, and processing specification information that specifies the processing in the processing block. The operation content type options are, for example, processing that uses a generation AI such as "multiple RAG action" or "LLM action," or other predetermined processing candidates. Information on the operation content setting items and operation content type options is stored in the storage unit 1012.

[0086] In steps S320 and S321, the processing information acquisition unit 10143 transmits the setting items of the operation content and the options of the type of operation processing to the user terminal 301. These processes are performed as a process of transmitting information on the setting screen of the operation content.

[0087] In step S322, the user inputs setting items and selects the type of operation processing based on the information acquired from the application management apparatus 101.

[0088] In steps S323 and S324, the processing information acquisition unit 10143 acquires the input setting items and the selected type of operation processing from the user terminal 301.

[0089] In step S325, the processing information acquisition unit 10143 stores the input information and the selection information as processing information.

[0090] 6 shows an example of an operation content setting screen 601. The operation content setting screen 601 has a common input area 6011, an output data structure input area 6012, a type selection area 6013, and an input area by type 6014.

[0091] In the common input area 6011, for example, the ID, name, and description of the processing block are input. In the example of FIG. 6, the processing block ID is "chat_node," the title is "Composite RAG·WEB," and the description is "Processing to execute Composite RAG·WEB." In the output data structure input area 6012, for example, the structure of data to be output by an operation is selected from structures set by the user and input. In the example of FIG. 6, a structure with a structure ID of "rag_content" and a structure with a structure ID of "answer" are selected. In the type selection area 6013, for example, the type of operation processing is selected from options displayed in a pull-down menu and input. In the type input area 6014, the content displayed varies depending on the type of operation processing selected in the type selection area 6013; for example, the name of the analysis model to be used and a prompt statement are input.

[0092] Returning to FIG. 3, in step S326, an operation is performed on the user terminal 301 to set transition conditions for application operations.

[0093] In step S327, the user terminal 301 requests the application management apparatus 101 to set a transition condition.

[0094] In step S328, the transition information acquisition unit 10144 generates a transition condition setting screen. The transition condition setting screen includes information for selecting a transition source processing block, information for selecting a transition destination processing block, and setting items for setting the transition conditions. Information on the setting items for the transition conditions is stored in the storage unit 1012.

[0095] In step S329, the transition information acquisition unit 10144 transmits information on the transition condition setting screen to the user terminal 301.

[0096] In step S330, the user inputs setting items based on the information acquired from the application management apparatus 101.

[0097] In step S331, the transition information acquisition unit 10144 acquires the input setting items from the user terminal 301.

[0098] In step S332, the transition information acquisition unit 10144 stores the input information from the user terminal 301 as a transition condition.

[0099] 7 shows an example of a transition condition setting screen 701. The transition condition setting screen 701 has a common input area 7011 and a state transition diagram 7012.

[0100] The common input area 7011 is used to input, for example, transition conditions from the start operation to the destination operation and an output message when the operation ends. In the example of FIG. 7, the source process block is specified as "Composite RAG·WEB" and the destination process block is specified as "END." Because answer generation by the "Composite RAG·WEB" process block is completed within the process block, no special conditions are set and the field is left blank. The message is set to display information corresponding to the structure with the structure ID "rag_content" and the structure ID "answer" when the process block transitions to "END." The state transition diagram 7012 visualizes the transition states of the process blocks. The state transition diagram 7012 visualizes the application's operation, making it easier for users to visualize the application's sequence of operations.

[0101] In step S333, an operation is performed on the user terminal 301 to set the second basic information of the application.

[0102] In step S334, the user terminal 301 requests the application management apparatus 101 to set second basic information including, for example, the structure of variable data used when starting the application and identifier information for authenticating or controlling access.

[0103] 3, in step S335, the second basic information acquisition unit 10145 generates a setting screen for the second basic information. Information on the setting items of the second basic information is stored in the storage unit 1012.

[0104] In step S336, the second basic information acquisition unit 10145 transmits to the user terminal 301 information on the setting screen for the second basic information of the application.

[0105] In step S337, the user inputs setting items via the second basic information setting screen.

[0106] In step S338, the second basic information acquisition unit 10145 acquires the input setting items from the user terminal 301.

[0107] In step S339, the second basic information acquisition unit 10145 stores the input information from the user terminal 301 as second basic information.

[0108] In step S340, a deployment operation is performed on the user terminal 301 to make the application available to the user.

[0109] In step S 341 , the user terminal 301 transmits deployment operation information to the application management apparatus 101 .

[0110] In step S342, the application management apparatus 101 stores the application as available for use by the user based on the deployment operation information.

[0111] The process described above allows application creators to set the processing in their applications through a simple operation of selecting from a predefined set of processing candidates. Furthermore, because the user can define the structure of the data processed in the application, it is possible to define the data structure and assign variable data to processing blocks in accordance with the user's needs. This allows applications using generative AI to be created with simple operations that reflect the user's needs.

[0112] The above-described embodiments are intended to facilitate understanding of the present invention and are not intended to limit the present invention. The elements of the embodiments, as well as their arrangement, conditions, shape, size, etc., are not limited to those exemplified and can be modified as appropriate. [Explanation of symbols]

[0113] 10...Application creation system, 101...Application management device, 201...Generation device, 301...User terminal, 1014...Application creation unit, 10142...Structural information acquisition unit, 10143...Processing information acquisition unit, 10144...Transition information acquisition unit, 2013...Analysis model unit

Claims

1. 1. A system for creating an application, comprising: a structure information acquisition unit that acquires structure information that defines the structure of a plurality of variable data items to be processed in the application; a processing information acquisition unit that acquires processing information defining a process in the application, the processing information including processing designation information that is assigned to each of a plurality of processing blocks and that designates a process selected from a plurality of pre-defined processing candidates including a process using a generation AI, and data designation information that designates variable data to be processed by each of the plurality of processing blocks as a variable having the structure defined by the structure information; a transition information acquisition unit that acquires transition information that defines the transition of processes between the plurality of processing blocks; A system comprising:

2. 10. The system of claim 1, the structural information includes explanatory information indicating a description of the structure; A system in which the explanatory information is included as input to the generating AI during processing using the generating AI.

3. 3. The system of claim 2, The system further comprises an explanation display unit that displays the explanation information to a user.

4. 10. The system of claim 1, The system includes processing using the generation AI including inputting a prompt to the generation AI requesting that the generation AI output structural correspondence information corresponding to the structure of a variable specified as an output for an AI processing block to which processing using the generation AI is assigned.

5. 5. The system of claim 4, The processing in the AI ​​processing block includes, when the structure correspondence information is not output from the generation AI, re-inputting the prompt to the generation AI; If the structural correspondence information is not output from the generation AI after a predetermined number of re-inputs, the system updates the variables designated as the output based on the information output from the generation AI.

6. 10. The system of claim 1, The structural information is First structure information indicating a structure of variable data preset in the application; and second structure information indicating a structure selected from the plurality of candidate variable data structures.

7. 1. A method of creating an application, comprising: The computer obtaining structure information defining a structure of a plurality of variable data to be processed in the application; Acquiring processing information that defines processing in the application, the processing information including processing designation information that is assigned to each of a plurality of processing blocks and that designates a processing selected from a plurality of pre-defined processing candidates including processing using a generation AI, and data designation information that designates variable data to be processed by each of the plurality of processing blocks as a variable having the structure defined by the structure information; obtaining transition information defining process transitions between the plurality of process blocks; How to do it.

8. A program for creating an application, On the computer, obtaining structure information defining a structure of a plurality of variable data to be processed in the application; Acquiring processing information that defines processing in the application, the processing information including processing designation information that is assigned to each of a plurality of processing blocks and that designates a processing selected from a plurality of pre-defined processing candidates including processing using a generation AI, and data designation information that designates variable data to be processed by each of the plurality of processing blocks as a variable having the structure defined by the structure information; obtaining transition information defining process transitions between the plurality of process blocks; A program that executes the following.

Citation Information

Patent Citations

  • Index-based parameter access and software to use it

    JP2008507755A

  • Program generating device, program generating method, and program

    JP2014078264A

  • Source code analyzing apparatus, source code analyzing method, and program

    JP2016076080A

  • New language conversion method in block type programming system

    JP2020161094A

  • System

    JP2025048865A