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

Abstract

This invention provides an information processing system, an information processing method, and an information processing program that can generate highly practical AI agents that respond to prompts. [Solution] The system comprises a storage means capable of storing management information relating to the management of a company; a learning means capable of learning at least the relationship (management information relationship) between one piece of management information and other pieces of management information; a prompt analysis means capable of analyzing a prompt input by a user and generating prompt analysis information including information on one or more tasks to realize the prompt; and an agent generation means capable of generating a workflow consisting of multiple nodes using the prompt analysis information as input and generating an AI agent to execute the workflow, wherein the prompt analysis means generates the prompt analysis information taking into account the management information relationships.

Description

Technical Field

[0001] The present invention relates to an information processing system, an information processing method, and an information processing program capable of generating a highly practical AI agent that conforms to a prompt.

Background Art

[0002] Conventionally, there has been proposed a business process creation support device including: a generation unit that receives an effect input by a user and a sentence indicating what is to be achieved, and generates a flow including a plurality of clauses obtained by morphological analysis of the sentence; a presentation unit that respectively obtains scores indicating which field each of the effect and the words included in the clause is close to, and presents a template of a business process flow in a field with a high total score to the user; and a creation support unit that supports the creation of a business process flow corresponding to the effect and the sentence (see, for example, Patent Document 1).

Prior Art Documents

Patent Documents

[0003]

Patent Document 1

Summary of the Invention

Problems to be Solved by the Invention

[0004] However, in the conventional device, when the information input by the user is insufficient, there is a problem that the degree of completion of the generated business process flow becomes low.

[0005] In view of such circumstances, the present invention aims to provide an information processing system, an information processing method, and an information processing program that can increase the degree of completion of an AI agent generated based on a workflow by enhancing the ability to analyze a prompt and increasing the degree of completion of the workflow generated based on the analysis result of the prompt, and can generate a highly practical AI agent that conforms to the prompt. [Means for solving the problem]

[0006] The information processing system according to the present invention comprises: a storage means capable of storing management information relating to the management of a company; a learning means capable of learning at least the relationship between one piece of management information and other pieces of management information (hereinafter referred to as "management information relationship"); a prompt analysis means capable of analyzing a prompt input by a user and generating prompt analysis information including information on one or more tasks to realize the prompt; and an agent generation means capable of generating a workflow consisting of multiple nodes using the prompt analysis information as input and generating an AI agent to execute the workflow, wherein the prompt analysis means generates the prompt analysis information taking into account the management information relationship.

[0007] The information processing method according to the present invention is an information processing method performed by an information processing system equipped with storage means capable of storing management information relating to the management of a company, wherein at least the following can be executed: a learning step capable of learning the relationship between one piece of management information and other pieces of management information (hereinafter referred to as "management information relationship"); a prompt analysis step capable of analyzing a prompt input by a user and generating prompt analysis information including information on one or more tasks to realize the prompt; and an agent generation step capable of generating a workflow consisting of multiple nodes using the prompt analysis information as input and generating an AI agent to execute the workflow, wherein the prompt analysis step generates the prompt analysis information taking into account the management information relationship.

[0008] The information processing program according to the present invention is an information processing program for an information processing system equipped with storage means capable of storing management information relating to the management of a company, wherein the computer is configured to function as a learning means capable of learning at least the relationship between one piece of management information and other pieces of management information (hereinafter referred to as "management information relationship"), a prompt analysis means capable of analyzing a prompt input by a user and generating prompt analysis information including information on one or more tasks to realize the prompt, and an agent generation means capable of generating a workflow consisting of multiple nodes using the prompt analysis information as input and generating an AI agent to execute the workflow, wherein the prompt analysis means generates the prompt analysis information taking into account the management information relationship. [Effects of the Invention]

[0009] According to the information processing system, information processing method, and information processing program of the present invention, the ability to analyze prompts is enhanced, and the completeness of the workflow generated based on the prompt analysis results is improved, thereby enhancing the completeness of the AI ​​agent generated based on the workflow and enabling the generation of a highly practical AI agent that conforms to the prompts, 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. [Figure 3] (a) A diagram showing an example of the relationship between tasks and management information for realizing a prompt. (b) A diagram showing an example of a workflow. [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 comprises: a storage means 12d capable of storing management information relating to the management of a company; a learning means 12a capable of learning at least the relationship between one piece of management information and other pieces of management information (management information relationship); a prompt analysis means 12b capable of analyzing a prompt input by a user and generating prompt analysis information including information on one or more tasks to realize the prompt; and an agent generation means 12c capable of generating a workflow consisting of multiple nodes using the prompt analysis information as input and generating an AI agent to execute the workflow, wherein the prompt analysis means 12b generates prompt analysis information taking into account the management information relationship.

[0014] According to the information processing system of this embodiment, by improving the ability to analyze prompts and improving the completeness of the workflow generated based on the prompt analysis results, the completeness of the AI ​​agent generated based on the workflow can be improved, and a highly practical AI agent that conforms to the prompt can be generated.

[0015] Here, "management information" refers to information related to the management of a company (for example, a retailer, a drugstore, or a convenience store). Examples of "management information" 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., market 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, and history of collaboration with headquarters).

[0016] "Management information relationships" refer to the relationship between one piece of management information and other pieces of management information. Examples of "management information relationships" include correlations between one piece of management information and other pieces of management information (for example, the relationship between a product's selling price and demand for that product, or the relationship between the location of a store selling a product and its sales), and causal relationships between one piece of management information and other pieces of management information (for example, the relationship between weather or temperature and product sales, or the relationship between sales promotion activities and product sales performance).

[0017] A "prompt" refers to instructions or tasks given by a user to a system. Examples include AI prompts given to a generating AI, and command prompts that provide instructions through command statements. The data format of a "prompt" is not particularly limited and may include text data, audio data, still image data, video data, or files.

[0018] A "task" refers to the operations and processes necessary to achieve a certain goal or objective. The "tasks" may be executed in parallel, in a predetermined order, according to a predetermined priority, or in a random order, either partially or entirely.

[0019] A "node" refers to the components that make up a workflow, and a workflow is composed of one or more nodes. Examples of "nodes" include a single process, a single judgment, a single branch, etc.

[0020] Note that the business information relationship may include the correlation or causal relationship between one piece of business information and another.

[0021] With such a configuration, the ability to analyze prompts can be further enhanced, and a highly practical AI agent compliant with the prompt can be generated.

[0022] In addition, business information includes purchase history information, which is information on the history of products purchased by the company's customers. The prompt is an issue related to the product, the prompt analysis means 12b is a large language model, and the agent generation means 12c may be an AI agent generation tool capable of generating an AI agent from a workflow.

[0023] With such a configuration, an AI agent for solving issues related to products can be generated based on natural language (text), enhancing the convenience for users.

[0024] Here, a Large-Scale Language Model (LLM) refers to an AI model that learns from vast amounts of text data and can understand and generate natural language (text) like a human. Examples of "Large-Scale Language Models" include AI models that perform tasks such as chatbots, search assistance and summarization, programming assistance such as code generation and debugging, business support such as email and report creation, question answering, and learning support.

[0025] Furthermore, the "task" according to the present invention may be a task related to management information or a task related to purchasing information.

[0026] "Purchasing information" refers to information such as product name, price, delivery date, and conditions that a company needs when purchasing goods from a supplier. Examples of "purchasing information" include purchase history information, product information (product ID, product name, category (JAN code), brand, unit price, etc.), order quantity, and delivery date.

[0027] "Purchase history information (data related to purchase history)" refers to information (past performance data) about the history of products purchased by a specific individual (customer). Examples of "purchase history information" include: (1) customer information (customer ID, gender, age, membership rank, etc.), (2) product information (product ID, product name, category (JAN code), brand, unit price, etc.), (3) purchase information (purchase date, purchase time, purchase quantity, total amount, payment method, surveys, staff interaction records, etc.), (4) store information (store ID, store name, sales channel (store / EC), etc.), (5) campaign information (status of campaigns and events, coupon usage, point usage, discount rate, etc.), and (6) purchase frequency (number of purchases, average purchase interval, most recent purchase date, etc.).

[0028] <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.

[0029] The information processing system 10 can be configured, for example, to include a system terminal 12 that controls the entire system, and an information storage terminal 12d and an external terminal 16 that are connected to the system terminal 12 via a network NW so that they can communicate with each other.

[0030] 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.

[0031] The information storage terminal 12d is a terminal (storage means) used by a company (for example, a drugstore, retail store, convenience store, etc.) to store the company's management information, and is composed of a POS terminal, a conventionally known server, a personal computer, etc.

[0032] As mentioned above, "management information" refers to information related to the management of a company (for example, a retail business or a drugstore). Examples of "management information" include: (1) sales and profit-related data (for example, sales data, profit data), (2) sales and product management-related data (for example, product master data and specifications, shelf layout data, sales performance and promotional effectiveness), (3) inventory and purchasing-related data (for example, inventory status, ordering and purchasing data), (4) customer and marketing-related data (for example, customer ID data (ID-POS), marketing campaign results), (5) store operation and personnel management-related data (for example, store performance, staff shifts), (6) expense and financial-related data (for example, store and headquarters expenses, financial indicators), (7) external environment data (for example, 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 (for example, work instructions and policies from superiors, reports and proposals from the field, and history of collaboration with headquarters).

[0033] In this example, the information storage terminal 12d is configured with a single server, but it may also be configured with multiple servers or personal computers. Furthermore, the terminal (storage means) for storing management information may be an internal storage means (for example, the storage device 26 shown in Figure 2) connected to the system terminal 12 via a local network (for example, LAN or P2P), or it may be the storage means of an external terminal 16, or it may be any other storage means.

[0034] External terminals 16 are terminals used by users of the information processing system 10 (for example, employees of a company), 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.

[0035] The network NW is a line that allows the system terminal 12, the information storage terminal 12d, and the 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.

[0036] <System Terminal / Hardware Configuration Example> Next, we will describe an example of the hardware configuration of system terminal 12.

[0037] 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.

[0038] 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.

[0039] 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.

[0040] 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.

[0041] 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, a plasma display, or an organic EL display. The communication unit 29 is a control means that controls communication with the information storage terminal 12d and external terminals 16, etc. via the network NW, and is composed of, for example, a communication card.

[0042] <System Terminal / Function> Next, we will explain the functions of the system terminal 12.

[0043] The storage device 26 of the system terminal 12 stores a program (information processing program) that causes the system terminal 12 to function as a learning means 12a, a prompt analysis means 12b, and an agent generation means 12c.

[0044] <System terminal / Functions / Learning methods> Next, we will explain the learning method 12a.

[0045] The learning means 12a is a means capable of learning at least the relationship between one piece of management information and other pieces of management information (management information relationship), and 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.

[0046] As mentioned above, "business information relationships" refer to the relationship between one piece of business information and other pieces of business information. Examples of "business information relationships" include correlations between one piece of business information and other pieces of business information (for example, the relationship between the selling price of a product and the demand for that product, or the relationship between the location of a store selling a product and the sales of that product), and causal relationships between one piece of business information and other pieces of business information (for example, the relationship between weather or temperature and product sales, or the relationship between sales promotion activities and product sales performance).

[0047] The learning means 12a periodically refers to multiple types of management information stored in the information storage terminal 12d and learns the relationship between one piece of management information and other pieces of management information (management information relationships).

[0048] For example, the learning means 12a refers to the sales price of a certain product included in the sales and product management related data of the management information and the sales performance information of a certain product included in the sales and product management related data of the management information. If it determines that there is a correlation between the sales price of a certain product and the sales performance of that product (for example, a correlation that lowering the sales price of a certain product makes that product sell well, or a correlation that raising the sales price of a certain product does not change the sales performance of that product), it stores these correlations in the storage device 26 as information indicating the relationship between management information.

[0049] Furthermore, the learning means 12a refers to the location of stores included in the expense and financial data of the management information and the sales data included in the sales and profit data of the management information, and if it determines that there is a correlation between the location of a store that sells a certain product and the sales of that product (for example, a correlation that sales of a certain product increase if the store that sells a certain product is located in an urban area, or a correlation that sales of a certain product decrease if the store that sells a certain product is located in a specific region compared to other regions), it stores these correlations in the storage device 26 as information indicating the relationship between management information.

[0050] Furthermore, the learning means 12a refers to temperature information included in the external environmental data of the management information and sales data included in the sales and profit-related data of the management information, and if it determines that there is a causal relationship between temperature and product sales (for example, a causal relationship that sales of a certain product decrease when the temperature drops, or a causal relationship that cold products sell well when the temperature rises), it stores these causal relationships in the storage device 26 as information indicating the relationship between management information.

[0051] Furthermore, the learning means 12a refers to the sales promotion activity information included in the sales and product management related data of the management information and the sales performance information of a certain product included in the sales and product management related data of the management information. If it determines that there is a causal relationship between the sales promotion activity and the product sales performance (for example, a causal relationship in which conducting sales promotion activities for a certain product leads to good sales of that product), it stores these causal relationships in the storage device 26 as information indicating the relationship between management information.

[0052] <System terminal / function / prompt analysis means> Next, the prompt analysis means 12b will be described.

[0053] The prompt analysis means 12b is a means capable of analyzing a prompt input by a user of the information processing system 10 and generating prompt analysis information, which includes information on one or more tasks to realize the prompt, taking into account the relationship between management information. In this example, it is composed of a program stored in the storage device 26 of the system terminal 12, and the storage device 26, etc.

[0054] Here, "prompt" refers to information (data) of tasks or instructions entered by a user of the information processing system 10 using an input means (keyboard, pointing device (mouse, etc.), touch panel, microphone, etc.) of an external terminal 16. The data format of the "prompt" is not particularly limited and may be any of the following: text data, audio data, still image data, video data, file, etc. Examples of "prompts" include data analysis, retrieval of specific information, creation and proofreading of documents, summarization of documents, code generation, document creation, and image generation.

[0055] The prompt analysis means 12b analyzes the prompt entered by a user using the external terminal 16 and generates prompt analysis information, which includes information on one or more tasks to realize the prompt, by adding information indicating the relationship between management information stored by the learning means 12a in the storage device 26.

[0056] For example, if an external terminal 16 provides a task (prompt) such as "I want to forecast the sales of product a," the prompt analysis means 12b generates a task to fulfill that task (forecast the sales of product a) based on the knowledge acquired through machine learning and the knowledge obtained from the external terminal 16.

[0057] The multiple tasks (processes) required to realize a sales forecast for product a include, for example, (Task 1) collecting sales performance data for product a (e.g., collecting POS data), (Task 2) collecting external factor data (weather, season, etc.), (Task 3) collecting market data (market trends, sales trends of similar products, etc.), (Task 4) collecting competitor data (prices of competing products, etc.), (Task 5) selecting a forecasting model (moving average method, exponential smoothing method, etc.), (Task 6) creating a sales forecast, and (Task 7) visualizing the sales forecast (creating a forecast graph, etc.).

[0058] As shown in Figure 3(a), the prompt analysis means 12b stores information on multiple tasks 1 to 7 for achieving the objective (sales forecast for product a) in the storage device 26, associating it with the user's ID.

[0059] Next, the prompt analysis means 12b refers to the information indicating the relationship between business information stored in the memory device 26 by the learning means 12a, and generates prompt analysis information taking this relationship between business information into account.

[0060] For example, consider a case where the memory device 26 stores information indicating a relationship between business information, such as "lowering the selling price of product a increases sales of product a."

[0061] In this case, the prompt analysis means 12b replaces the processing content of "(Task 6) Create sales forecast" with the processing content of "Create a sales forecast for product a, taking into account the correlation that lowering the selling price of product a will increase sales of product a," as shown in Figure 3(a). This allows the processing content of "(Task 6) Create sales forecast" to be updated to match the user's prompt, thereby improving the ability to analyze prompts.

[0062] Furthermore, consider a case where, as information indicating a relationship between business information, the memory device 26 stores a correlation such as, "Sales of product a increase when the store selling product a is located in an urban area."

[0063] In this case, the prompt analysis means 12b adds information such as "Sales of product a increase when the store selling product a is located in an urban area" to "(Process 4) Collection of competitive data (prices of competing products, etc.)" as shown in Figure 3(a). This increases the amount of data that can be collected by the task "(Process 4) Collection of competitive data (prices of competing products, etc.)", and because the data is backed by management information, the validity of the data can be enhanced, and the content of the task for realizing the prompt can be further enriched.

[0064] Furthermore, consider a case where, for example, information indicating a relationship between management information, such as the causal relationship "sales of product a decrease when the temperature drops," is stored in memory device 26.

[0065] In this case, the prompt analysis means 12b adds the information "Sales of product a decrease when the temperature drops" to "(Process 2) Collection of external factor data (weather, season, etc.)" as shown in Figure 3(a). This increases the amount of data that can be collected by the task "(Process 2) Collection of external factor data (weather, season, etc.)", and because the data is backed by management information, its effectiveness can be enhanced, and the content of the task for realizing the prompt can be further enriched.

[0066] Furthermore, consider a case where, for example, information indicating a relationship between management information, such as the causal relationship "If we conduct sales promotion activities for product a, product a will sell well," is stored in memory device 26.

[0067] In this case, the prompt analysis means 12b adds information such as "product a will sell well if promotional activities for product a are carried out" to "(Process 3) Collection of market data (market trends, sales trends of similar products, etc.)" as shown in Figure 3(a). This increases the amount of data that can be collected by the task "(Process 3) Collection of market data (market trends, sales trends of similar products, etc.)", and because the data is backed by management information, the validity of the data can be enhanced, and the content of the task for realizing the prompt can be further enriched.

[0068] Next, the prompt analysis means 12b stores information obtained by adding information indicating the relationship between management information to the information of multiple tasks 1 to 7 for achieving the task (sales forecast of product a), as prompt analysis information, associated with the user's ID, in the storage device 26.

[0069] The prompt analysis means 12b may be configured to generate and output code (program code) for the agent generation means 12c to generate a workflow corresponding to the prompt analysis information, and to import said code into the agent generation means 12c.

[0070] With this configuration, the code generated and output by the prompt analysis means 12b can be used in existing workflow creation support tools, etc., thereby improving user convenience.

[0071] <System terminal / function / agent generation method> Next, the agent generation means 12c will be described.

[0072] The agent generation means 12c is a means capable of generating a workflow consisting of multiple nodes using prompt analysis information as input, and generating an AI agent (program) that executes the workflow. In this example, it is composed of a program stored in the storage device 26 of the system terminal 12, and the storage device 26, etc.

[0073] As mentioned above, a "node" is a component (part) that makes up a workflow, and a workflow consists of one or more nodes. Examples of nodes include a single process, a single decision, a single branch, etc.

[0074] For example, consider a case where the prompt analysis information includes the information for tasks 1 to 7 shown in Figure 3(a), and information indicating the relationship with management information, which is stored in the memory device 26.

[0075] In this case, the agent generation means 12c generates a workflow consisting of multiple nodes ND1 to ND13, for example, as shown in Figure 3(b).

[0076] Node ND1 is the node (process) corresponding to "(Task 1) Collection of sales performance data for product a," and for example, it executes the task of collecting sales performance data for product a from the management information stored in the information storage terminal 12d. Node ND2 is the node (determination) that determines whether or not Task 1 has been completed; if Task 1 has not been completed, it returns to node ND1, and if Task 1 has been completed, it proceeds to node ND3.

[0077] Node ND3 is the node (process) corresponding to "(Task 2) Collection of external factor data." For example, it retrieves information stored as information indicating business information relationships, such as "Sales of product a decrease when the temperature drops," and also performs the task of collecting external factor data from external websites, etc., via the network NW. Node ND4 is the node (determination) that determines whether or not Task 2 has been completed. If Task 2 has not been completed, it returns to node ND3; if Task 2 has been completed, it proceeds to node ND5.

[0078] Node ND5 is the node (process) corresponding to "(Task 3) Market Data Collection." For example, it retrieves information stored as information indicating management information relationships, such as "Lowering the selling price of product a will increase sales of product a," and also performs the task of collecting market data from external websites, etc., via the network NW. Node ND6 is the node (determination) that determines whether or not Task 3 has been completed. If Task 3 has not been completed, it returns to node ND5; if Task 3 has been completed, it proceeds to node ND7.

[0079] Node ND7 is the node (process) corresponding to "(Task 4) Collection of Competitive Data." For example, it retrieves information stored as information indicating business information relationships, such as "Sales of product a increase when the store selling product a is located in an urban area," and also performs the task of collecting competitive data from external websites, etc., via the network NW. Node ND8 is the node (determination) that determines whether or not Task 4 has been completed. If Task 4 has not been completed, it returns to node ND7; if Task 4 has been completed, it proceeds to node ND9.

[0080] Node ND9 is the node (process) corresponding to "(Task 5) Selection of a prediction model," and it performs the selection of a prediction model. Node ND10 is a node (determination) that determines whether or not Task 5 has been completed. If Task 5 has not been completed, it returns to node ND9; if Task 5 has been completed, it proceeds to node ND11.

[0081] Node ND11 is the node (process) corresponding to "(Task 6) Creating a Sales Forecast," and executes the task of "creating a sales forecast for product a" by referring to information showing the relationship between management information and considering the correlation that "lowering the selling price of product a will increase sales of product a." Node ND12 is the node (determination) that determines whether or not Task 6 has been completed; if Task 6 has not been completed, it returns to node ND11, and if Task 6 has been completed, it proceeds to node ND13. Node ND13 is the node (process) corresponding to "(Task 7) Visualizing the Sales Forecast (Creating a Forecast Graph, etc.)," ​​and executes the creation of a forecast graph, etc.

[0082] Next, the agent generation means 12c generates an AI agent (program) to execute the workflow based on the created workflow, and sends the AI ​​agent (program) to the external terminal 16 in response to a request from the external terminal 16.

[0083] Users of the information processing system 10 can use the received AI agent (program) to obtain output (for example, a graph of sales forecasts for product a) that corresponds to the prompt they input to the information processing system 10 (for example, a request to "predict sales of product a").

[0084] As this example demonstrates, by improving the ability to analyze prompts and enhancing the completeness of the workflows generated based on the prompt analysis results, it is possible to improve the completeness of the AI ​​agents generated based on the workflows, thereby generating highly practical AI agents that are in line with the prompts.

[0085] <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 characterized by comprising: a storage means capable of storing management information relating to the management of a company (for example, a storage means 12d shown in Figure 1, an information storage terminal 12d shown in Figure 2); a learning means capable of learning at least the relationship between one piece of management information and other pieces of management information (hereinafter referred to as "management information relationship") (for example, a learning means 12a shown in Figures 1 and 2); a prompt analysis means capable of analyzing a prompt input by a user and generating prompt analysis information including information on one or more tasks to realize the prompt (for example, a prompt analysis means 12b shown in Figures 1 and 2); and an agent generation means capable of generating a workflow consisting of multiple nodes using the prompt analysis information as input and generating an AI agent to execute the workflow, wherein the prompt analysis means generates the prompt analysis information taking into account the management information relationship.

[0086] 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 by an information processing system (for example, the information processing system 10 shown in Figures 1 and 2) equipped with storage means (for example, storage means 12d shown in Figure 1, information storage terminal 12d shown in Figure 2) capable of storing management information relating to the management of a company, and includes a learning step (for example, the method executed by the learning means 12a shown in Figures 1 and 2) capable of learning at least the relationship between one piece of management information and other pieces of management information (hereinafter referred to as "management information relationship"), and analyzes the prompts input by the user, and the program The information processing method is characterized in that it is at least capable of performing a prompt analysis step (for example, a method performed by prompt analysis means 12b shown in Figures 1 and 2) that generates prompt analysis information including information on one or more tasks for realizing an application, and an agent generation step (for example, a method performed by agent generation means 12c shown in Figures 1 and 2) that generates a workflow consisting of multiple nodes using the prompt analysis information as input and generates an AI agent that executes the workflow, wherein the prompt analysis step generates the prompt analysis information taking into account the management information relationships.

[0087] Furthermore, the information processing program according to this embodiment (for example, a program executed by the information processing system 10 shown in Figures 1 and 2) is an information processing program for an information processing system (for example, the information processing system 10 shown in Figures 1 and 2) equipped with storage means (for example, storage means 12d shown in Figure 1, information storage terminal 12d shown in Figure 2) capable of storing management information relating to the management of a company, and the computer is equipped with learning means (for example, learning means 12a shown in Figures 1 and 2) capable of learning at least the relationship between one piece of management information and other pieces of management information (hereinafter referred to as "management information relationship"), and input by the user The information processing program is characterized by comprising: a prompt analysis means (for example, prompt analysis means 12b shown in Figures 1 and 2) capable of analyzing a prompt and generating prompt analysis information containing information on one or more tasks to realize the prompt; and an agent generation means (for example, agent generation means 12c shown in Figures 1 and 2) capable of generating a workflow consisting of multiple nodes using the prompt analysis information as input and generating an AI agent to execute the workflow, wherein the prompt analysis means generates the prompt analysis information taking into account the management information relationships.

[0088] According to the information processing system, information processing method, and information processing program of this embodiment, the ability to analyze prompts is enhanced, and the completeness of the workflow generated based on the prompt analysis results is improved, thereby increasing the completeness of the AI ​​agent generated based on the workflow and enabling the generation of a highly practical AI agent that conforms to prompts.

[0089] Furthermore, the aforementioned relationship of management information may include a correlation or causal relationship between the first piece of management information and the other piece of management information.

[0090] This configuration further enhances the ability to analyze prompts, enabling the creation of highly practical AI agents that adhere to the prompt-based programming conventions.

[0091] Furthermore, the management information includes purchase history information, which is information about the history of products purchased by the company's customers; the prompt is a problem related to the product; the prompt analysis means is a large-scale language model; and the agent generation means may be an AI agent generation tool capable of generating the AI ​​agent from the workflow.

[0092] With this configuration, it becomes possible to generate AI agents that solve product-related problems based on natural language (text), thereby improving user convenience.

[0093] Furthermore, the prompt analysis means may be capable of outputting the prompt analysis information as a code, and the agent generation means may be capable of generating the AI ​​agent using the code as input.

[0094] With this configuration, the code generated and output by the prompt analysis means can be used in existing workflow creation support tools, etc., thereby improving user convenience.

[0095] It should be noted that the information processing system, information processing method, information processing program, and AI agent 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.

[0096] Therefore, for example, the "correlation" according to the present invention is not limited to the relationship between the selling price of a product and the demand for the product, or the relationship between the location of the store selling the product and the sales of the product, but may also be the relationship between advertising expenses and sales, the location of the store selling the product and the number of customers, etc.

[0097] Furthermore, the "causal relationship" according to the present invention is not limited to the relationship between weather or temperature and product sales, or the relationship between sales promotion activities and product sales performance, but may also refer to the relationship between the layout of the store where the product is sold and product sales performance, or the relationship between the frequency of coupon distribution and the number of customers, etc. [Industrial applicability]

[0098] The information processing system, information processing method, and information processing program according to the present invention can be widely applied to fields such as manufacturing, service industries, and retail industries. [Explanation of Symbols]

[0099] 10. Information Processing Systems 12 System Terminals 12a Learning methods 12b Prompt analysis means 12c Agent generation means 12d Information Storage Terminal 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. A storage means capable of storing management information related to the management of a company, A learning means capable of learning at least the relationship between one piece of management information and other pieces of management information (hereinafter referred to as "management information relationship"), A prompt analysis means capable of analyzing a prompt entered by a user and generating prompt analysis information that includes information on one or more tasks to fulfill the prompt, The system includes an agent generation means capable of generating a workflow consisting of multiple nodes using the aforementioned prompt analysis information as input, and generating an AI agent that executes the workflow, The prompt analysis means generates the prompt analysis information taking into account the management information relationships. An information processing system characterized by the following:

2. In the information processing system described in claim 1, The aforementioned relationship of management information includes the correlation or causal relationship between one piece of management information and the other piece of management information. An information processing system characterized by the following:

3. In the information processing system according to claim 1 or 2, The aforementioned management information includes purchase history information, which is information about the history of products purchased by the company's customers. The aforementioned prompt is a problem relating to the aforementioned product, The prompt analysis means is a large-scale language model, The agent generation means is an AI agent generation tool capable of generating the AI ​​agent from the workflow. An information processing system characterized by the following:

4. In the information processing system according to claim 1 or 2, The prompt analysis means is capable of outputting the prompt analysis information as a code. The agent generation means is capable of generating the AI ​​agent using the code as input. An information processing system characterized by the following:

5. An information processing method performed by an information processing system equipped with storage means capable of storing management information relating to the management of a company, A learning step that enables learning at least the relationship between one piece of management information and other pieces of management information (hereinafter referred to as "management information relationship"), A prompt analysis step capable of analyzing a prompt entered by a user and generating prompt analysis information that includes information on one or more tasks to fulfill the prompt, At least one agent generation step is possible, which involves generating a workflow consisting of multiple nodes using the aforementioned prompt analysis information as input, and generating an AI agent capable of executing the workflow. The prompt analysis step generates the prompt analysis information taking into account the management information relationships. An information processing method characterized by the following:

6. In the information processing method described in claim 5, The aforementioned relationship of management information includes the correlation or causal relationship between one piece of management information and the other piece of management information. An information processing method characterized by the following:

7. An information processing program for an information processing system equipped with a storage means capable of storing management information relating to the management of a company, Computers, A learning means capable of learning at least the relationship between one piece of management information and other pieces of management information (hereinafter referred to as "management information relationship"), A prompt analysis means capable of analyzing a prompt entered by a user and generating prompt analysis information that includes information on one or more tasks to fulfill the prompt, The agent generation means is capable of generating an AI agent that generates a workflow consisting of multiple nodes using the aforementioned prompt analysis information as input, and then generates an AI agent that executes the workflow. The prompt analysis means generates the prompt analysis information taking into account the management information relationships. An information processing program characterized by the following features.

8. In the information processing program described in claim 7, The aforementioned relationship of management information includes the correlation or causal relationship between one piece of management information and the other piece of management information. An information processing program characterized by the following features.

Images