Neural Network Action Sequence Segmentation for Automation Discovery
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current methods for discovering and analyzing business processes for automation are manual, time-consuming, prone to mistakes, and biased, often missing significant automation opportunities and requiring high skill levels, which makes them costly and inefficient.
Innovation Solution
A system and method that uses a neural network to segment computer-based actions by determining loss thresholds, allowing for the identification of significant business flows for automation by breaking down sequences of actions into segments, thereby improving the discovery of automation opportunities.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If manual discovery and analysis of business processes is performed, then high skill level analysts can identify automation opportunities, but the process is time-consuming, expensive, and prone to mistakes
Solution Approach 1:
The patent replaces the manual mechanical analysis process performed by human analysts with an automated machine learning system. The system uses supervised learning models to automatically discover business processes and identify automation opportunities from user action logs, eliminating the need for time-consuming manual analysis while maintaining or improving accuracy.
Solution Approach 2:
The patent introduces machine learning models as intermediaries between raw user action data and automation opportunity identification. These models process and interpret user behavior patterns, transforming unstructured logs into structured process discoveries that would otherwise require manual analysis by skilled analysts.
2Adaptability or versatility
If manual analysis is used to identify business processes, then analysts can understand what to automate, but the discovery process is biased and significant flows can be easily missed
Solution Approach 1:
The patent replaces subjective human judgment with objective machine learning algorithms that systematically analyze user action data. The supervised learning models provide unbiased, consistent identification of business processes without the cognitive biases that affect manual analysis, ensuring more reliable and complete discovery of automation opportunities.
3Ease of manufacture
If skilled analysts perform process discovery manually, then automation sequences can be created with deep understanding, but the process is very expensive and requires high skill levels
Solution Approach 1:
The patent replaces the need for highly skilled manual analysts with automated machine learning systems that perform process discovery. This substitution democratizes the capability, allowing organizations to perform comprehensive process analysis without requiring expensive expert resources, thereby easing the creation of automation sequences.
Solution Approach 2:
The patent enables the system to automatically perform process discovery and identification tasks that previously required skilled human analysts. The machine learning models self-train on user action data and autonomously identify business processes, eliminating the need for external expert intervention and reducing operational complexity.
Data Source
AI summary
A system and method for segmenting or dividing a series of computer-based actions, for example into sentences, may provide a sequence of subsets of the series of actions to a neural network using a sliding window, and divide or segment the series actions into segments at points where the loss of the neural network is above a threshold. The dividing may include, for each of a sequence of computer-based actions within a sliding window determining if the sequence when provided to the neural network corresponds to a loss above or equal to a threshold, and if so, determining that an action in the sequence of actions within the sliding window should not be part of a segment or sentence being created.