Semantic Action Graphs for Accurate RPA Task Determination
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing robotic process automation (RPA) technologies fail to efficiently identify and automate repetitive tasks due to inefficiencies in task mining and process determination, leading to reduced productivity.
Innovation Solution
The use of semantic action graphs generated through task mining, normalization, classification, clustering, reinforcement learning, and supervised learning to determine and recommend actions, enhancing RPA efficiency.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If traditional task mining methods are used to identify repetitive tasks, then task identification can be performed, but the efficiency and accuracy are insufficient leading to reduced productivity
Solution Approach 1:
The patent segments the task identification process into multiple specialized components: normalization module (standardizing task data), classification module (categorizing tasks using classification algorithms), clustering module (grouping similar tasks using clustering algorithms), and reinforcement learning module (optimizing task identification). This segmentation allows each module to specialize in one aspect, improving overall accuracy and productivity compared to traditional single-method approaches.
Solution Approach 2:
The patent introduces semantic action graphs as an intermediary representation that connects raw task mining data to automated process recommendations. These graphs serve as a mediator that transforms unstructured task data into structured, actionable insights, enabling more accurate task identification and automation opportunities detection while improving productivity.
2Measurement precision
If multiple processing steps (normalization, classification, clustering, reinforcement learning) are applied to task mining data, then task identification accuracy improves, but system complexity increases
Solution Approach 1:
The patent divides the complex processing system into distinct, modular components: normalization module, classification module, clustering module, and reinforcement learning module. Each module performs a specific function and can be independently developed, tested, and maintained. This segmentation reduces system complexity by making the overall system more manageable while still achieving high task identification accuracy through the coordinated work of all modules.
Solution Approach 2:
The patent applies normalization as a preliminary action before classification and clustering. By standardizing the task mining data first, the subsequent modules work with pre-processed, consistent data, which simplifies their operations and reduces the complexity of handling raw, unstructured data. This preliminary processing step makes the overall system more efficient and manageable.
3Reliability
If semantic action graphs are generated through multiple learning processes, then RPA automation accuracy improves, but processing time and computational resources increase
Solution Approach 1:
The patent performs normalization, classification, and clustering as preliminary actions before generating semantic action graphs and applying reinforcement learning. This staged approach allows the system to prepare and organize data in advance, making the final graph generation and learning processes more efficient. By doing preliminary work upfront, the system reduces the computational burden and processing time during actual RPA automation execution, while maintaining high accuracy.
Data Source
Figure 1
Figure 2
Figure 3
AI summary
Action and/or process determination and recommendations for Robotic Process Automation (RPA) using semantic action graphs is disclosed. Semantic action graphs are graphs that store individual actions, and potentially graphical elements and/or text associated with the actions, as nodes, as well as the relationships between nodes as edges. Metadata to develop the semantic action graphs may be derived from task mining applications that can monitor the interactions of users with computing systems, workforce intelligence, etc. The semantic action graphs may be for a user, an organization, an industry, product-wide, etc. At their lowest level of granularity, the recommendations may be for mouse clicks, key presses, Application Programming Interface (API) calls, system events, etc. At higher levels of granularity, the recommendations may be for opening an order, creating a lead, approving a work item, etc.