Listener-Guided RPA Robot Personalization From User Interactions
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current robotic process automation (RPA) robots are not adequately adapted to the specific needs of individual users, leading to inefficiencies and performance issues, as they are often trained on general functionality rather than user-specific tasks.
Innovation Solution
A system that includes a user computing system with an RPA robot and a listener, which monitors user interactions, logs data, and transmits it to a server for analysis. The server determines if modifications are needed to the RPA workflow and inserts activities or sequences of activities to personalize the robot's functionality based on user-specific data, using both global and local machine learning models to improve performance.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Adaptability or versatility
If RPA robots are designed based on anticipated general functionality needs, then they can perform a wide range of tasks, but they are not adapted to the specific needs of individual users
Solution Approach 1:
The RPA robot performs self-training by automatically monitoring its own user interactions through the listener, logging data about user corrections and modifications, and using this data to generate updated workflows without requiring external reprogramming. This enables the robot to adapt to user-specific needs autonomously.
Solution Approach 2:
The system implements a feedback loop where the listener continuously monitors user interactions with the RPA robot, captures corrections or modifications made by the user, and feeds this information back to the robot. The robot then uses this feedback to learn and adjust its behavior, improving user-specific adaptation over time.
2Productivity
If RPA robots are trained on general functionality, then deployment is faster, but performance on user-specific tasks is insufficient
Solution Approach 1:
The system performs preliminary training actions during the robot's initial deployment with general functionality. The listener begins monitoring user interactions immediately, and the robot starts learning user-specific patterns concurrently with normal operation, rather than requiring a separate training phase that would delay deployment.
Solution Approach 2:
The robot continuously learns and improves its performance throughout operation rather than completing training in a discrete phase. The listener continuously monitors interactions, and the robot continuously updates its workflows based on accumulated data, ensuring productive action is maintained while learning occurs.
3Adaptability or versatility
If the system monitors and logs all user interactions, then personalized adaptation is achieved, but data processing load increases
Solution Approach 1:
The listener extracts only the specific data elements needed for training from user interactions, such as corrections, modifications, and exception handling patterns. Rather than processing all interaction data, the system selectively captures relevant training information, reducing computational load while maintaining personalization capability.
Data Source
AI summary
Human-in-the-loop robot training using artificial intelligence (AI) for robotic process automation (RPA) is disclosed. This may be accomplished by a listener robot watching interactions of a user or another robot with a computing system. Based on the interactions by the user or robot with the computing system, the robot may be improved and/or personalized for the user or a group of users.