Cognitive Agent Process Learning Graph Generation
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Cognitive agents lack knowledge of process steps, making it difficult for them to provide support for processes, as they typically rely on subject matter experts to convey this information into a knowledge base, but there is no easy method for experts to do so.
Innovation Solution
A method and system for generating a process learning graph and document output from a recorded process, where a subject matter expert executes the process, records it, and the system augments the recording to create a knowledge base for the cognitive agent, including screen captures, system responses, and user input requirements, allowing the agent to understand and support process execution.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Loss of information
If subject matter experts manually convey process information into a knowledge base, then the cognitive agent gains process knowledge, but the complexity and time required for knowledge base creation increases
Solution Approach 1:
The system automatically records and captures the expert's process execution as it occurs, creating a knowledge base entry that is a direct copy of the actual process demonstration rather than requiring manual transcription or documentation
Solution Approach 2:
The system performs preliminary actions by automatically capturing screen captures, system responses, and user inputs during the expert's process execution, preparing the knowledge base data before the expert completes the process demonstration
2Loss of information
If subject matter experts manually convey process information into a knowledge base, then the cognitive agent gains process knowledge, but the time required for knowledge base creation increases
Solution Approach 1:
The system performs preliminary actions by automatically capturing screen captures, system responses, and user inputs during the expert's process execution, preparing the knowledge base data before the expert completes the process demonstration
Solution Approach 2:
The system captures the expert's process execution in real-time as it occurs, creating knowledge base entries simultaneously with the process demonstration rather than requiring separate documentation time
3Ease of operation
If the cognitive agent lacks process step knowledge, then the system architecture remains simple, but the agent's ability to provide process support deteriorates
Solution Approach 1:
The system automatically records and captures the expert's process execution as it occurs, creating a knowledge base entry that is a direct copy of the actual process demonstration rather than requiring manual transcription or documentation
Solution Approach 2:
The system acts as an intermediary between the subject matter expert and the cognitive agent, automatically capturing and structuring process information from the expert's demonstration and delivering it to the agent in a usable format
Data Source
AI summary
One embodiment provides a method for generating a process learning graph and a document output from a recorded process for training a cognitive agent, the method comprising: utilizing at least one processor to execute computer code that performs the steps of: obtaining a recording of a process, wherein the recording comprises a demonstration of executing the process; generating, using the recording, the process learning graph, wherein the process learning graph identifies a process flow; generating, using the recording, the document output, wherein the document output comprises process screen transitions and process steps; and providing the process learning graph and the document output to the cognitive agent. Other aspects are described and claimed.