ML Graph for Automated Software Test Configuration
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Software testing configurations face inefficiencies due to redundant test cases and test steps, which are time and resource intensive to identify and manage, especially as projects evolve, leading to increased storage requirements and computational waste.
Innovation Solution
A graph data structure is generated to represent unique test steps, with unsupervised and supervised machine learning models used to identify redundancies and predict impacts of changes, allowing for automated management and optimization of test cases and test steps.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If manual identification and management of test cases is used, then testing coverage can be maintained, but time and resource consumption increases significantly
Solution Approach 1:
The patent replaces manual mechanical identification and management of test cases with automated machine learning models. The supervised ML model automatically identifies impacted test cases when project requirements change, eliminating the need for manual analysis while maintaining comprehensive testing coverage.
Solution Approach 2:
The system enables self-service automation where the ML models autonomously perform test case identification, redundancy detection, and impact analysis without human intervention. The unsupervised model self-organizes test cases into clusters and identifies redundancies, while the supervised model self-adapts to project changes and predicts impacted test cases.
2Reliability
If comprehensive test suites are created to ensure coverage, then testing reliability improves, but storage requirements and computational resources increase
Solution Approach 1:
The unsupervised machine learning model extracts and identifies redundant test cases from the comprehensive test suite by analyzing similarities in test steps, inputs, and expected outputs. These redundant cases are then removed or consolidated, reducing storage requirements while preserving essential testing coverage.
Solution Approach 2:
The system changes the parameter of test case representation by creating condensed versions of redundant test cases or merging them into single representative test cases. This transformation reduces the quantity of stored test data while maintaining the ability to achieve comprehensive coverage through strategic selection of representative cases.
3Measurement precision
If detailed tracking of test case impacts is implemented, then change management accuracy improves, but system complexity increases
Solution Approach 1:
The patent replaces complex manual tracking systems with machine learning-based prediction models. The supervised ML model automatically predicts which test cases are impacted by project changes by learning patterns from historical data, achieving high accuracy without requiring complex manual tracking infrastructure.
4Loss of substance
If redundant test cases are manually identified and removed, then storage efficiency improves, but productivity decreases due to time-consuming analysis
Solution Approach 1:
The unsupervised machine learning model performs self-service redundancy detection by automatically analyzing test case similarities, clustering redundant cases, and identifying candidates for removal. This automation eliminates the need for manual analysis while improving productivity through rapid, scalable processing of large test suites.
Solution Approach 2:
The patent substitutes manual redundancy identification with automated ML-based analysis that rapidly processes test cases to detect redundancies. This mechanical substitution dramatically improves productivity by handling large volumes of test cases in minutes rather than hours or days of manual work.
Data Source
AI summary
Various embodiments are directed to performing automated software testing configuration management operations. In one embodiment, a method is provided. The method includes generating a graph data structure including nodes representing test steps across test cases of a software testing data object for a software application platform and including segments each connecting two nodes and representing test cases having the two represented test steps. The method further includes receiving a query describing a testing change and generating extracted features from the query. The method further includes generating test outcome requirements and graph portions of the graph data structure relevant to the testing change using supervised machine learning models and the extracted features. The method further includes generating, in response to the query, an updated graph data structure and a change report indicating the relevant test outcome requirements and performing prediction-based actions using the updated graph data structure and the change report.