Information processing system, information processing method, and information processing program

Abstract

This invention provides an information processing system, method, and program that can identify the processes that a system under analysis can execute, and that can improve the development efficiency of the system by utilizing or applying the processes of that system. [Solution] An information processing system 10 is configured to have an AI agent 12, wherein the AI ​​agent includes a learning means 12a capable of learning a combination of input data and output data generated by executing a specific process based on the input data as learning data, an analysis means 12b capable of acquiring output data generated by inputting input data into a process that the system to be analyzed can execute, and an analysis means 12b that refers to the output data to be analyzed and the output data of the learning data, and when the output data to be analyzed matches or approximates the output data of the learning data, it estimates that the process to be analyzed that outputs the output data to be analyzed is a process that has the same function as a specific process.

Description

Technical Field

[0005] ,

[0001] The present invention relates to an information processing system, an information processing method, and an information processing program capable of grasping processes executable by a system to be analyzed.

Background Art

[0002] Conventionally, means for solving problems of existing systems using generative AI, and systems including system migration services, reverse engineering services, system performance improvement, refactoring services, programming language conversion services, and customization function services have been proposed (see, for example, Patent Document 1). [[ID=​​​​​​​​​​​​​​​​​​​​​​​​​​​​​​​The information processing system according to the present invention is an information processing system configured to have an AI agent, wherein the AI ​​agent comprises: a learning means capable of learning a combination of input data and output data generated by executing a specific process based on the input data as learning data; an analysis means capable of acquiring output data (hereinafter referred to as "analyzed output data") generated by inputting the input data to a process that the system to be analyzed can execute (hereinafter referred to as "analyzed process"); and a verification means that refers to the analyzed output data and the output data of the learning data and performs a verification process to verify whether the analyzed output data matches or approximates the output data of the learning data, wherein the verification means estimates that the analyzed process that outputs the analyzed output data is a process that has the same function as the specific process, when the analyzed output data matches or approximates the output data of the learning data.

[0007] The information processing method according to the present invention is It is a computer AI Agent by execution It will be done An information processing method, characterized in that the AI ​​agent performs at least the following steps: a learning step in which it can learn a combination of input data and output data generated by performing a specific process based on the input data as learning data; a data acquisition step in which it can acquire output data (hereinafter referred to as "analyzed output data") generated by inputting the input data into a process that the system to be analyzed can execute (hereinafter referred to as "analyzed process"); and a verification step in which it performs a verification process that refers to the analyzed output data and the output data of the learning data and verifies whether the analyzed output data matches or approximates the output data of the learning data, wherein the verification step estimates that the analyzed process that outputs the analyzed output data is a process that has the same function as the specific process.

[0008] The information processing program according to the present invention is a program for an information processing system configured to have an AI agent, wherein the AI ​​agent is configured to function as a learning means capable of learning a combination of input data and output data generated by executing a specific process based on the input data as learning data; an analysis means capable of acquiring output data (hereinafter referred to as "analyzed output data") generated by inputting the input data into a process that the system to be analyzed can execute (hereinafter referred to as "analyzed process"); and a verification means that references the analyzed output data and the output data of the learning data and performs a verification process to verify whether the analyzed output data matches or approximates the output data of the learning data, wherein the verification means estimates that the analyzed process that outputs the analyzed output data is a process that has the same function as the specific process. [Effects of the Invention]

[0009] According to the information processing system, information processing method, and information processing program of the present invention, it is possible to understand the processing that the system to be analyzed can perform, and by utilizing or applying the processing of the system, the development efficiency of the system can be improved, which is an excellent effect. [Brief explanation of the drawing]

[0010] [Figure 1] This is a schematic diagram showing an overview of the information processing system 10 according to this embodiment. [Figure 2] This is a system configuration diagram showing an example of the configuration of the information processing system 10 according to this embodiment. [Modes for carrying out the invention]

[0011] The following describes an information processing system 10 according to an embodiment of the present invention with reference to the drawings.

[0012] <Overview of the Information Processing System> First, an overview of the information processing system 10 according to this embodiment will be described using Figure 1. Figure 1 is a schematic diagram showing an overview of the information processing system 10 according to this embodiment.

[0013] The information processing system 10 according to this embodiment is an information processing system configured to have an AI agent 12, wherein the AI ​​agent 12 comprises: a learning means 12a capable of learning a combination of input data and output data generated by executing a specific process based on the input data as learning data; an analysis means 12b capable of acquiring output data (analyzed output data) generated by inputting input data to a process that the system to be analyzed can execute (analyzed process); and a verification means 12c that refers to the analyzed output data and the output data of the learning data and executes a verification process to verify whether the analyzed output data matches or approximates the output data of the learning data, wherein the verification means 12c estimates that the analyzed process that outputs the analyzed output data is a process that has the same function as a specific process when the analyzed output data matches or approximates the output data of the learning data.

[0014] According to the information processing system 10 of this embodiment, it is possible to understand the processes that the system to be analyzed can execute, and the development efficiency of the system can be improved by utilizing or applying the processes of the system.

[0015] Here, "AI agent" refers to a system that acts as a substitute (agent) for a human, possessing the ability to learn and make decisions on its own, and integrating various AI technologies for solving complex problems.

[0016] AI agents can perform tasks based on external triggers (e.g., user prompts, sensor inputs) or internal triggers (e.g., activation by an internal scheduler, occurrence of anomalies).

[0017] For example, when a question (prompt) is given as a trigger from the outside, tasks such as searching for an answer to the question or generating an answer are executed, and the answer to the question is output. Also, for example, when a task (prompt) is given as a trigger from the outside, tasks such as searching for a solution to the task or generating a solution are executed, and the solution to the task is output.

[0018] Also, "prompt" refers to an instruction or question given by a user to the system, and includes, for example, an AI prompt given to a generative AI, a command prompt for giving an instruction by a command sentence (command), etc. The data format of the "prompt" is not particularly limited, and may be any of, for example, text data, voice data, still image data, moving image data, files, etc.

[0019] Various processes executed by the AI agent can also be executed by "agent-type AI", "optical generation model", etc. Here, "agent-type AI" refers to a system in which a plurality of AI agents operate in cooperation. Agent-type AI is used to solve complex problems that cannot be solved by a single AI agent. Each AI agent plays a specialized role and outputs an answer to a question or a solution to a task while exchanging information with each other.

[0020] Also, "optical generation model" refers to a system that uses the physical properties of light and combines optical elements such as lenses, prisms, and diffraction gratings so that light itself performs calculations and processes information. Since it can process countless light rays simultaneously, its processing speed is higher than that of an AI that uses electricity to process information sequentially. Also, its energy efficiency is high, and it can reduce power consumption compared to an AI that uses electricity.

[0021] "Specific process" refers to a process that can be executed by the AI agent 12 and that generates (outputs) output data based on input data.

[0022] As the "specific process", for example, when a prompt (input data) is input, tasks such as searching for an answer to the prompt or generating an answer are executed, and large language (LLM) processing capable of generating (outputting) an answer (output data) to the prompt; when data related to business numerical values (input data) is input, business analysis processing capable of executing business analysis and generating (outputting) data related to business analysis (output data); when data indicating an instruction to generate an image (input data) is input, generative AI processing capable of generating an image and generating (outputting) image data (output data), etc. can be cited. Note that the data format of the "input data" input to the specific process and the "output data" generated by the specific process is not particularly limited, and may be any of, for example, text data, voice data, still image data, moving image data, files, etc.

[0023] The "system" refers to a mechanism in which a plurality of hardware and software with predetermined functions are related to each other and function as a single unit, and which is capable of generating (outputting) output data (analysis target output data) by inputting input data.

[0024] Examples of "systems" include: a large-scale language (LLM) system that, when given a prompt (input data), performs tasks such as searching for or generating an answer to the prompt, and generates (outputs) the answer to the prompt (output data to be analyzed); a business analysis system that, when given data related to business figures (input data), performs business analysis and generates (outputs) data related to the business analysis (output data to be analyzed); and a generative AI system that, when given data indicating an instruction to generate an image (input data), generates an image and generates (outputs) image data (output data to be analyzed). The data format of the "input data" that is input to the processes that the system can execute, and the "output data (output data to be analyzed)" that the processes that the system can execute generate, are not particularly limited and may be any of the following: text data, audio data, still image data, video data, files, etc.

[0025] "Data related to management figures" refers to numerical information (data) that visualizes a company's management status through management indicators and financial indicators, and is used to evaluate company performance. Examples of "data related to management figures" include: (1) sales and profit-related data (e.g., sales data, profit data), (2) sales and product management-related data (e.g., product master data, specifications, shelf layout data, sales performance and promotional effectiveness), (3) inventory and purchasing-related data (e.g., inventory status, ordering and purchasing data), (4) customer and marketing-related data (e.g., customer ID data (ID-POS), marketing campaign results), (5) store operations and personnel management-related data (e.g., store performance, staff shifts), (6) expense and financial-related data (e.g., store and headquarters expenses, financial indicators), (7) external environment data (e.g., trade area / population data, competitor store location and pricing information, weather, temperature, and disaster information), (8) local event / school event information, and (9) instruction and communication-related data (e.g., work instructions and policies from superiors, reports and suggestions from the field, history of collaboration with headquarters).

[0026] Furthermore, within "management numerical information," (2) sales and product management-related data (e.g., product master data, specifications, shelf layout data, sales performance and promotional effects) may be referred to as "sales promotion data," and within "management numerical data," (4) customer and marketing-related data (e.g., customer ID data (ID-POS), marketing campaign results) may be referred to as "purchase history data (purchase history information)."

[0027] The verification means 12c may store at least one or more processes executed in the verification process until the output data to be analyzed matches or approximates the output data of the training data, the logic or algorithm used in said one or more processes, and the data used in said one or more processes.

[0028] With this configuration, the system can store (learn) the processes, logic, algorithms, and data until the output data to be analyzed matches or approximates the output data of the training data. Furthermore, the learned processes, logic, algorithms, and data can be used to support the validation process, and new processes, logic, algorithms, and data can be generated based on the learned processes, logic, algorithms, and data.

[0029] Furthermore, the learning means 12a may be capable of learning multiple types of learning data patterns where the input data is the same but the specific processing and output data are different, and the verification means 12c may change the pattern of the learning data being referenced and continue the verification process if the output data to be analyzed does not match or approximate the output data of the learning data.

[0030] With this configuration, the system can perform analysis multiple times based on training data of multiple patterns, thereby enhancing the system's analytical capabilities.

[0031] Furthermore, the input data may be prompts, the system may be a large-scale language system, and the output data of the data to be analyzed and the output data of the training data may be responses to prompts.

[0032] With this configuration, it becomes possible to understand the processing capabilities of the large-scale language system being analyzed, and by utilizing or applying the processing capabilities of that large-scale language system, the efficiency of system development can be improved.

[0033] Furthermore, the input data may be data relating to business figures, the system may be a business analysis system that performs business analysis, and the output data of the analysis target and the training data may be data that shows the results of the business analysis.

[0034] With this configuration, it becomes possible to understand the processes that the management analysis system being analyzed can perform, and the development efficiency of the system can be improved by utilizing or adapting the processes of that management analysis system.

[0035] Furthermore, the learning means 12a can learn information about the analysis target process that the verification means 12c has estimated to have the same function as a specific process (for example, the function of the analysis target process, the storage location of the analysis target process (for example, URL)), and the AI ​​agent 12 may generate output data by inputting the input data to the system's analysis target process instead of executing a specific process when given input data.

[0036] With this configuration, even if a specific process is not implemented in the AI ​​agent, it becomes possible to generate output data by inputting the data into the analysis target process of an external system, thereby improving the efficiency of system development.

[0037] <Example of system configuration> Next, an example of the configuration of the information processing system 10 according to this embodiment will be described using Figure 2. Figure 2 is a system configuration diagram showing an example of the configuration of the information processing system 10 according to this embodiment.

[0038] The information processing system 10 can be configured, for example, to include a system terminal 12 that controls the entire system, and external systems 14 and 16 that are connected to the system terminal 12 via a network NW so that they can communicate with each other.

[0039] The system terminal 12 is a terminal that controls the entire information processing system 10, and is composed of conventionally known servers, personal computers, etc. In this example, the system terminal 12 is composed of one server, but it may be composed of multiple servers, personal computers, etc. The hardware configuration of the system terminal 12 and the programs that the system terminal 12 executes will be described later.

[0040] External system 14 is one, multiple, or all of the following systems: a large-scale language (LLM) system, a business analysis system, a generative AI system, etc., and is composed of conventionally known servers, personal computers, etc. In this example, external system 14 is configured with one server, but it may be configured with multiple servers, personal computers, etc.

[0041] External terminals 16 are terminals used by users of the information processing system 10, and consist of personal computers, tablets, smartphones, etc. The type of external terminal 16 is not particularly limited, but examples include smartphones, personal computers, tablets, etc. used by individuals.

[0042] The network NW is a line that allows system terminal 12, external system 14, and external terminal 16 to communicate with each other, and is typically composed of a WAN (Wide Area Network), also known as the Internet. The network NW may be wired or wireless, a LAN (Local Area Network), a dedicated line such as a VPN (Virtual Private Network), or a combination of these lines.

[0043] <System Terminal / Hardware Configuration Example> Next, we will describe an example of the hardware configuration of system terminal 12.

[0044] As shown in Figure 2, the system terminal 12 is configured to include, for example, a CPU 21, and a ROM 22, RAM 23, external storage drive 25, storage device 26, input device 27, display device 28, communication unit 29, etc., all connected to the CPU 21 via a bus.

[0045] The CPU 21 is a control means that controls the entire system terminal 12, and performs processes such as executing application programs and operating systems (OS) stored in ROM 22 and storage devices 26, and storing data and files necessary for program execution in RAM 23 and storage devices 26.

[0046] ROM22 is a storage means for storing basic I / O programs and various data, and is composed of, for example, PROM, flash memory, etc. RAM23 is a storage means for temporarily storing data, and is composed of, for example, SDRAM, DRAM, etc. External storage drive25 is a control means that can read and write data to recording media 24 such as magnetic tape, DVD, etc., and is composed of, for example, magnetic tape storage, DVD drive, etc.

[0047] The storage device 26 is a storage means for storing application programs, the OS, control programs, related programs, various information, etc., and is composed of, for example, a hard disk drive (HDD), a solid-state drive (SDD), etc. The input device 27 is for inputting commands (instructions), etc., to the system terminal 12, and is composed of, for example, a keyboard, a pointing device (mouse, etc.), a touch panel, etc.

[0048] The display device 28 displays commands input by the input device 27, the response output of the system terminal 12 to those commands, and various other displays, and is composed of, for example, a liquid crystal display, plasma display, organic EL, etc. The communication unit 29 is a control means that controls communication with external systems 14 and external terminals 16, etc. via the network NW, and is composed of, for example, a communication card, etc.

[0049] <System Terminal / Function> Next, we will explain the functions of the system terminal 12.

[0050] The storage device 26 of the system terminal 12 stores a program (information processing program) that enables the system terminal 12 to function as a learning means 12a, an analysis means 12b, and a verification means 12c.

[0051] <System terminal / Functions / Learning methods> Next, we will explain the learning method 12a.

[0052] The learning means 12a is a means capable of learning a combination of input data and output data generated by performing a specific process based on the input data as learning data. In this example, it is composed of a program stored in the storage device 26 of the system terminal 12, or the storage device 26, etc.

[0053] As stated above, "specific processing" refers to processing that the AI ​​agent 12 can execute and that generates (outputs) output data based on input data.

[0054] Examples of "specific processes" include large-scale language (LLM) processing that, when a prompt (input data) is input, performs tasks such as searching for an answer to the prompt or generating an answer, and generates (outputs) an answer to the prompt (output data); business analysis processing that, when data related to business figures (input data) is input, performs business analysis and generates (outputs) data related to the business analysis (output data); and generative AI processing that, when data indicating an instruction to generate an image (input data) is input, generates an image and generates (outputs) image data (output data). The data format of the "input data" input to a specific process and the "output data" generated by a specific process is not particularly limited and may be any of the following: text data, audio data, still image data, video data, files, etc.

[0055] When the AI ​​agent 12 performs a specific process based on the input data and generates output data, the learning means 12a associates the input data, the output data, and information about the specific process and stores them in the memory device 26.

[0056] For example, if the AI ​​agent 12 performs large-scale language (LLM) processing based on a prompt (e.g., a question consisting of text data) given from outside the system terminal 12 (e.g., an external terminal 16) and generates an answer to the prompt (e.g., an answer consisting of text data), the learning means 12a associates the prompt (input data), the answer to the prompt (output data), and information about the specific processing (large-scale language processing) (e.g., the content of the processing and the algorithm) and stores them in the memory device 26.

[0057] Furthermore, for example, if the AI ​​agent 12 performs a business analysis process based on business performance data (e.g., current period sales data composed of files) provided from within the system terminal 12 (e.g., the CPU 21), and generates business analysis data (e.g., a graph showing the current period's sales trend composed of image data, and analysis results composed of text data), the learning means 12a associates the business performance data (input data), the business analysis data (output data), and information related to a specific process (business analysis process) (e.g., the content of the process and the algorithm) and stores them in the storage device 26.

[0058] Furthermore, for example, if the AI ​​agent 12 executes an image generation process based on data indicating an instruction to generate a specific image given from outside the system terminal 12 (for example, an external terminal 16) (for example, a command consisting of text data), and generates image data (for example, image data generated based on the command), the learning means 12a associates the data indicating an instruction to generate a specific image (input data), the image data (output data), and information related to a specific process (business analysis process) (for example, the content of the process and the algorithm) and stores them in the storage device 26.

[0059] <System terminal / Function / Analysis method> Next, the analytical means 12b will be described.

[0060] The analysis means 12b is a means capable of acquiring output data (analyzed output data) generated by inputting input data into a process (analyzed process) that the system to be analyzed can execute. In this example, it consists of a program stored in the storage device 26 of the system terminal 12, and the storage device 26, etc.

[0061] As described above, a "system" is a mechanism in which multiple pieces of hardware and software with predetermined functions interact with each other and function as a single unit, capable of generating (outputting) output data (output data to be analyzed) by inputting input data.

[0062] Examples of "systems" include: a large-scale language (LLM) system that, when given a prompt (input data), performs tasks such as searching for or generating an answer to the prompt, and generates (outputs) the answer to the prompt (output data to be analyzed); a business analysis system that, when given data related to business figures (input data), performs business analysis and generates (outputs) data related to the business analysis (output data to be analyzed); and a generative AI system that, when given data indicating an instruction to generate an image (input data), generates an image and generates (outputs) image data (output data to be analyzed). The data format of the "input data" that is input to the processes that the system can execute, and the "output data (output data to be analyzed)" that the processes that the system can execute generate, are not particularly limited and may be any of the following: text data, audio data, still image data, video data, files, etc.

[0063] The analysis means 12b inputs input data into a process that the system to be analyzed can execute (analysis target process) and obtains output data (analysis target output data) output from the system.

[0064] For example, consider a case where the system to be analyzed (e.g., external system 14) is a large-scale language system that, when a prompt (input data) is input, performs tasks such as searching for an answer to the prompt or generating an answer, and is capable of generating (outputting) an answer to the prompt (output data to be analyzed).

[0065] In this case, the analysis means 12b inputs, for example, a prompt (for example, a question consisting of text data) as input data to the large-scale language system, obtains a response to the prompt (for example, a response consisting of text data) from the large-scale language system, and stores the obtained response in the storage device 26 as output data to be analyzed by the large-scale language system.

[0066] Furthermore, consider the case where the system to be analyzed (for example, external system 14) is a business analysis system that, when data related to business figures (input data) is input, performs business analysis and generates (outputs) data related to business analysis (output data to be analyzed).

[0067] In this case, the analysis means 12b inputs data relating to business figures (for example, sales data for the current period composed of files) as input data to the business analysis system, and acquires data relating to business analysis (for example, a graph showing the sales trend for the current period composed of image data and analysis results composed of text data) from the business analysis system, and stores the acquired data relating to business analysis in the storage device 26 as the output data to be analyzed by the business analysis system.

[0068] Furthermore, consider the case where the system to be analyzed (for example, external system 14) is a generative AI system that, when it receives data indicating an instruction to generate an image (certain input data), generates an image and generates (outputs) image data (output data to be analyzed).

[0069] In this case, the analysis means 12b inputs, for example, data indicating an instruction to generate a specific image (for example, a command consisting of text data) as input data to the generation AI system, and acquires generated image data (for example, image data generated based on the command) from the generation AI system, and stores the acquired image data in the storage device 26 as output data to be analyzed by the generation AI system.

[0070] <System terminal / Function / Verification method> Next, the verification means 12c will be described.

[0071] Verification means 12c refers to the output data to be analyzed and the output data of the training data, performs a verification process to verify whether the output data to be analyzed matches or approximates the output data of the training data, and if the output data to be analyzed matches or approximates the output data of the training data, it is a means to estimate that the processing to be analyzed that outputs the output data to be analyzed is a processing that has the same function as a specific processing. In this example, it is composed of a program stored in the storage device 26 of the system terminal 12, or the storage device 26, etc.

[0072] For example, consider a case where the memory device 26 stores the answers to prompts (e.g., answers consisting of text data) as output data to be analyzed, and the memory device 26 also stores the answers to prompts (e.g., answers consisting of text data) as output data for training.

[0073] In this case, the verification means 12c refers to the output data to be analyzed and the output data of the training data, and performs a verification process to verify whether the output data to be analyzed matches or approximates the output data of the training data. If the output data to be analyzed matches or approximates the output data of the training data, it estimates that the analysis target process that outputs the analysis target data (in this example, a large-scale language processing process that can be executed by the external system 14) is a process that has the same function as a specific process (in this example, a large-scale language processing process that can be executed by the AI ​​agent 12).

[0074] Furthermore, consider a case where, for example, the memory device 26 stores data related to business analysis (for example, a graph showing the sales trend for the current period, consisting of image data, and the analysis results, consisting of text data) as output data for analysis, and the memory device 26 also stores data related to business analysis (for example, a graph showing the sales trend for the current period, consisting of image data, and the analysis results, consisting of text data) as output data for learning.

[0075] In this case, the verification means 12c refers to the output data to be analyzed and the output data of the training data, and performs a verification process to verify whether the output data to be analyzed matches or approximates the output data of the training data. If the output data to be analyzed matches or approximates the output data of the training data, it estimates that the analysis target process that outputs the output data to be analyzed (in this example, a business analysis process that can be executed by the external system 14) is a process that has the same function as a specific process (in this example, a business analysis process that can be executed by the AI ​​agent 12).

[0076] Furthermore, consider a case where, for example, the memory device 26 stores image data (for example, image data generated based on a command) as output data to be analyzed, and image data (for example, image data generated based on a command) as output data for training.

[0077] In this case, the verification means 12c refers to the output data to be analyzed and the output data of the training data, and performs a verification process to verify whether the output data to be analyzed matches or approximates the output data of the training data. If the output data to be analyzed matches or approximates the output data of the training data, it estimates that the analysis target process that outputs the analysis target data (in this example, an image generation process that can be executed by the external system 14) is a process that has the same function as a specific process (in this example, an image generation process that can be executed by the AI ​​agent 12).

[0078] In this example, it is possible to understand the processes that the system being analyzed can perform, and by utilizing or applying the processes of that system, the efficiency of system development can be improved.

[0079] Furthermore, the learning means 12a can learn multiple types of learning data patterns where the input data is the same but the specific processing and output data are different, and the verification means 12c may change the pattern of the learning data being referenced and continue the verification process if the output data to be analyzed does not match or approximate the output data of the learning data.

[0080] With this configuration, the system can perform analysis multiple times based on training data of multiple patterns, thereby enhancing the system's analytical capabilities.

[0081] Furthermore, the learning means 12a can learn information about the analysis target process that the verification means 12c has estimated to have the same function as a specific process, and the AI ​​agent 12 may generate output data by inputting the input data to the system's analysis target process instead of executing a specific process when given input data.

[0082] With this configuration, even if a specific process is not implemented in the AI ​​agent, it becomes possible to generate output data by inputting the data into the analysis target process of an external system, thereby improving the efficiency of system development.

[0083] <Information Processing Systems / Summary> As described above, the information processing system according to this embodiment (for example, the information processing system 10 shown in Figures 1 and 2) is an information processing system configured to have an AI agent (for example, the AI ​​agent 12 shown in Figure 1, the system terminal 12 shown in Figure 2), and the AI ​​agent has a learning means (for example, the learning means 12a shown in Figures 1 and 2) that can learn a combination of input data and output data generated by executing a specific process based on the input data as learning data, and an output data generated by inputting the input data into a process that the system to be analyzed can execute (hereinafter referred to as the "analysis target process") The information processing system comprises an analysis means (for example, the analysis means 12b shown in Figures 1 and 2) capable of acquiring data (hereinafter referred to as "analysis target output data"), and a verification means (for example, the verification means 12c shown in Figures 1 and 2) that references the analysis target output data and the output data of the learning data and performs a verification process to verify whether the analysis target output data matches or approximates the output data of the learning data, wherein the verification means estimates that the analysis target process that outputs the analysis target output data when the analysis target output data matches or approximates the output data of the learning data is a process that has the same function as a specific process.

[0084] Furthermore, the information processing method according to this embodiment (for example, the method executed by the information processing system 10 shown in Figures 1 and 2) is an information processing method executed using an AI agent (for example, the AI ​​agent 12 shown in Figure 1, the system terminal 12 shown in Figure 2), wherein the AI ​​agent can learn a combination of input data and output data generated by executing a specific process based on the input data as learning data (for example, the process executed by the learning means 12a shown in Figures 1 and 2), and the output data generated by inputting the input data to a process that the system to be analyzed can execute (hereinafter referred to as the "analysis target process") (hereinafter referred to as the "analysis target process"). The information processing method is characterized by performing at least the following steps: a data acquisition step (for example, a process performed by the analysis means 12b shown in Figures 1 and 2) that can obtain the output data to be analyzed (referred to as "evidence output data"), and a verification step (for example, a process performed by the verification means 12c shown in Figures 1 and 2) that refers to the output data to be analyzed and the output data of the learning data and performs a verification process to verify whether the output data to be analyzed matches or approximates the output data of the learning data, wherein the verification means estimates that the analysis target process that outputs the output data to be analyzed is a process that has the same function as a specific process when the output data to be analyzed matches or approximates the output data of the learning data.

[0085] Furthermore, the information processing program according to this embodiment (for example, the program executed by the information processing system 10 shown in Figures 1 and 2) includes the AI ​​agent (for example, the AI ​​agent 12 shown in Figure 1, the system terminal 12 shown in Figure 2), and the AI ​​agent includes a learning means (for example, the learning means 12a shown in Figures 1 and 2) that can learn a combination of input data and output data generated by executing a specific process based on the input data as learning data, and output data generated by inputting the input data into a process that the system to be analyzed can execute (hereinafter referred to as the "analysis target process") (hereinafter referred to as the "minute processing"). This information processing program is characterized by comprising an analysis means (for example, analysis means 12b shown in Figures 1 and 2) capable of obtaining the "output data to be analyzed," and a verification means (for example, verification means 12c shown in Figures 1 and 2) that references the output data to be analyzed and the output data of the learning data and performs a verification process to verify whether the output data to be analyzed matches or approximates the output data of the learning data, wherein the verification means estimates that the analysis target process that outputs the output data to be analyzed is a process that has the same function as a specific process when the output data to be analyzed matches or approximates the output data of the learning data.

[0086] According to the information processing system, information processing method, and information processing program of this embodiment, it is possible to understand the processing that the system to be analyzed can perform, and the development efficiency of the system can be improved by utilizing or applying the processing of the system.

[0087] The verification means may also store at least one or more processes performed in the verification process until the output data to be analyzed matches or approximates the output data of the training data, the logic or algorithm used in the one or more processes, and the data used in the one or more processes.

[0088] With this configuration, the system can store (learn) the processing, logic, algorithms, and data until the output data to be analyzed matches or approximates the output data of the training data. Furthermore, the learned processing, logic, algorithms, and data can be used to support the validation process. Additionally, new processing, logic, algorithms, and data can be generated based on the learned processing, logic, algorithms, and data.

[0089] Furthermore, the learning means may be capable of learning multiple types of learning data in which the input data is the same but the specific processing and output data are different, and the verification means may change the learning data being referenced and continue the verification process if the output data to be analyzed does not match or approximate the output data of the learning data.

[0090] With this configuration, the system can perform analysis based on training data of multiple patterns, thereby enhancing its analytical capabilities.

[0091] Furthermore, the input data may be a prompt, the system may be a large-scale language system, and the output data of the analyzed output data and the training data may be responses to the prompt.

[0092] With this configuration, it becomes possible to understand the processing capabilities of the large-scale language system being analyzed, and by utilizing or applying the processing capabilities of that large-scale language system, the efficiency of system development can be improved.

[0093] Furthermore, the input data may be data relating to management figures, the system may be a management analysis system that performs management analysis, and the output data of the analysis target and the learning data may be data indicating the results of the management analysis.

[0094] With this configuration, it becomes possible to understand the processes that the management analysis system being analyzed can perform, and the development efficiency of the system can be improved by utilizing or adapting the processes of that management analysis system.

[0095] Furthermore, the learning means is capable of learning information about the analysis target process that the verification means has estimated to have the same function as the specific process, and the AI ​​agent may generate the output data by inputting the input data to the analysis target process of the system instead of executing the specific process when given the input data.

[0096] With this configuration, even if a specific process is not implemented in the AI ​​agent, it becomes possible to generate output data by inputting the data into the analysis target process of an external system, thereby improving the efficiency of system development.

[0097] It should be noted that the information processing system, information processing method, and information processing program according to the present invention are not limited to the embodiments described above, and various modifications can be made without departing from the spirit of the present invention.

[0098] Therefore, for example, the "specific processing" according to the present invention is not limited to large-scale language (LLM) processing, business analysis processing, or generative AI processing, but may be any processing that the AI ​​agent 12 can execute and that generates (outputs) output data based on input data, such as an automatic ordering process that automatically places orders for products.

[0099] Furthermore, the "system" according to the present invention is not limited to large-scale language (LLM) systems, business analysis systems, or generative AI systems, but can be any system capable of generating (outputting) output data (analyzable output data) by inputting input data, for example, an automated ordering system that automatically places orders for goods. [Industrial applicability]

[0100] The information processing system, information processing method, and information processing program according to the present invention can be widely applied to fields such as the retail industry. [Explanation of symbols]

[0101] 10 Information Processing Systems 12 System Terminals 12a Learning methods 12b Analysis tools 12c Verification method 14 External Systems 16 External terminals 21 CPU 22 ROM 23 RAM 24 Recording media 25 External storage drives 26 Storage device 27 Input devices 28 Display device 29 Communications Department

Claims

1. An information processing system configured to include an AI agent, The aforementioned AI agent, A learning means capable of learning a combination of input data and output data generated by performing a specific process based on said input data, An analysis means capable of obtaining output data (hereinafter referred to as "analyzed output data") generated by inputting the input data into a process that the system to be analyzed can execute (hereinafter referred to as "analyzed process"), The system includes a verification means that performs a verification process to verify whether the output data to be analyzed matches or approximates the output data of the training data, by referring to the output data to be analyzed and the output data of the training data. The verification means estimates that the analysis target process that outputs the analysis target data is a process having the same function as the specific process, when the analysis target output data matches or approximates the training data output data. An information processing system characterized by the following:

2. In the information processing system described in claim 1, The verification means stores at least one or more processes performed in the verification process until the output data to be analyzed matches or approximates the output data of the training data, the logic or algorithm used in the one or more processes, and the data used in the one or more processes. An information processing system characterized by the following:

3. In the information processing system according to claim 1 or 2, The learning means is capable of learning multiple types of learning data patterns in which the input data is the same but the specific processing and output data are different. The verification means continues the verification process by changing the pattern of the training data being referenced if the output data to be analyzed does not match or approximate the output data of the training data. An information processing system characterized by the following:

4. In the information processing system according to claim 1 or 2, The aforementioned input data is a prompt, The aforementioned system is a large-scale language system, The output data of the data to be analyzed and the output data of the training data are responses to the prompt. An information processing system characterized by the following:

5. In the information processing system according to claim 1 or 2, The aforementioned input data is data relating to business figures, The aforementioned system is a business analysis system that performs business analysis, The output data of the data to be analyzed and the output data of the learning data are data that show the results of the business analysis. An information processing system characterized by the following:

6. In the information processing system according to claim 1 or 2, The learning means is capable of learning information about the analysis target process that has been estimated by the verification means to be a process having the same function as the specific process, The AI ​​agent, when given the input data, can generate the output data by inputting the input data into the system's analysis target process instead of executing the specific process. An information processing system characterized by the following:

7. An information processing method performed by an AI agent which is a computer, The aforementioned AI agent, A learning step that can learn using a combination of input data and output data generated by performing a specific process based on the input data as training data, A data acquisition step that allows acquisition of output data (hereinafter referred to as "analyzed output data") generated by inputting the aforementioned input data into a process that the system to be analyzed can execute (hereinafter referred to as "analyzed process"), At least one verification step is performed, which involves referencing the output data to be analyzed and the output data of the training data, and performing a verification process to verify whether the output data to be analyzed matches or approximates the output data of the training data. The verification step, if the output data to be analyzed matches or approximates the output data of the training data, estimates that the analysis process that outputs the output data to be analyzed is a process that has the same function as the specific process. An information processing method characterized by the following:

8. In the information processing method described in claim 7, The learning step is capable of learning multiple types of learning data patterns in which the input data is the same but the specific processing and output data are different. The verification step continues by changing the pattern of the training data being referenced if the output data to be analyzed does not match or approximate the output data of the training data. An information processing method characterized by the following:

9. In the information processing method according to claim 7 or 8, The learning step is capable of learning information about the analysis target process that has been estimated by the verification step to have the same function as the specific process, The AI ​​agent, when given the input data, can generate the output data by inputting the input data into the system's analysis target process instead of executing the specific process. An information processing method characterized by the following:

10. A program for an information processing system configured to include an AI agent, The aforementioned AI agent, A learning means capable of learning a combination of input data and output data generated by performing a specific process based on said input data, An analysis means capable of obtaining output data (hereinafter referred to as "analyzed output data") generated by inputting the input data into a process that the system to be analyzed can execute (hereinafter referred to as "analyzed process"), This system functions as a verification means that performs a verification process to verify whether the output data to be analyzed matches or approximates the output data of the training data, by referring to the output data to be analyzed and the output data of the training data. The verification means estimates that the analysis target process that outputs the analysis target data is a process having the same function as the specific process, when the analysis target output data matches or approximates the training data output data. An information processing program characterized by the following features.

11. In the information processing program described in claim 10, The learning means is capable of learning multiple types of learning data patterns in which the input data is the same but the specific processing and output data are different. The verification means continues the verification process by changing the pattern of the training data being referenced if the output data to be analyzed does not match or approximate the output data of the training data. An information processing program characterized by the following features.

12. In the information processing program according to claim 10 or 11, The learning means is capable of learning information about the analysis target process that has been estimated by the verification means to be a process having the same function as the specific process, The AI ​​agent, when given the input data, can generate the output data by inputting the input data into the system's analysis target process instead of executing the specific process. An information processing program characterized by the following features.

Images