Information processing systems, information processing methods, and information processing programs

Abstract

It possesses: a memory mechanism for memorizing business information related to the company's operations; a learning mechanism for learning the relationship (business information relationship) between at least one piece of prior business information and other prior business information; a prompting and parsing mechanism for parsing prompts input by the user and generating prompting and parsing information including information on one or more tasks required to achieve the prompt; and an agent generation mechanism for generating a workflow consisting of multiple nodes using prior prompting and parsing information as input and generating an AI agent to execute the workflow; the prior prompting and parsing mechanism takes into account the relationship between prior business information when generating prior prompting and parsing information.

Claims

1. An information processing system, characterized by comprising: a memory means for memorizing business information concerning the operation of an enterprise; a learning means for learning the relationship between at least one piece of prior business information and other prior business information (hereinafter referred to as "business information relationship"); a prompting parsing means for parsing prompts input by a user and generating prompting parsing information including information on one or more tasks required to achieve the prompt; and an agent generation means for generating a workflow consisting of multiple nodes by taking prior prompting parsing information as input and generating an AI agent to execute the workflow; the prior prompting parsing means generates prior prompting parsing information by taking prior business information relationship into consideration.

2. The information processing system as described in claim 1, wherein, The relational relationship of the preceding business information includes: the correlation or causal relationship between a piece of the preceding business information and other pieces of the preceding business information.

3. The information processing system as described in request item 1 or 2, wherein, The preceding business information includes: information on the purchase history of the products purchased by the preceding enterprise's customers, i.e., purchase history information; preceding keywords, which are topics about the preceding products; preceding keyword parsing method, which is a large-scale language model; preceding agent generation method, which is an AI agent generation tool that can generate preceding AI agents from the preceding workflow.

4. The information processing system as described in request item 1 or 2, wherein, The pre-recorded teleprompter parsing method can output pre-recorded teleprompter parsing information as code; the pre-recorded agent generation method can use the pre-recorded code as input to generate a pre-recorded AI agent.

5. An information processing method, which is an information processing method executed by an information processing system having a memory means for memorizing business information related to business operations, characterized in that it can at least execute: a learning step, which can learn the relationship between at least one piece of prior business information and other prior business information (hereinafter referred to as "business information relationship"); and a prompt parsing step, which can parse the prompts input by the user and generate prompt parsing information including information on one or more tasks required to realize the prompts; and an agent generation step, which can generate a workflow consisting of multiple nodes by taking the prior prompt parsing information as input and generate an AI agent to execute the workflow; the prior prompt parsing step generates the prior prompt parsing information by taking the relationship of prior business information into consideration.

6. The information processing method as described in request item 5, wherein, The relational relationship of the preceding business information includes: the correlation or causal relationship between a piece of the preceding business information and other pieces of the preceding business information.

7. An information processing program, which is an information processing system having a memory means for memorizing business information related to business operations, characterized in that the computer functions as follows: a learning means, which can learn the relationship between at least one piece of prior business information and other prior business information (hereinafter referred to as "business information relationship"); a prompting parsing means, which can parse prompts input by users and generate prompting parsing information including information on one or more tasks required to achieve the prompt; and an agent generation means, which can use prior prompting parsing information as input to generate a workflow consisting of multiple nodes and generate an AI agent that executes the workflow; the prior prompting parsing means generates prior prompting parsing information by taking prior business information relationship into consideration.

8. The information processing program as described in request item 7, wherein, The relational relationship of the preceding business information includes: the correlation or causal relationship between a piece of the preceding business information and other pieces of the preceding business information.

Images