Defect Management Engine for Software Test Coverage
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current software development processes often fail to detect all defects in software products due to insufficient test cases and test coverage, leading to costly and time-consuming defect fixing, involving multiple teams and affecting customer satisfaction and organizational losses.
Innovation Solution
A defect management engine creates a master list of features from a code repository, maps defects to the most probable features, labels unmapped defects as 'orphan defects', and calculates test coverage and sufficiency metrics to identify gaps in testing, using techniques like unsupervised extraction and distance metrics.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If traditional manual defect management processes are used, then defect detection relies on human effort and multiple teams, but this leads to incomplete defect detection, high costs, and time-consuming resolution
Solution Approach 1:
The patent replaces manual mechanical defect management processes with an automated defect management engine that uses machine learning models, natural language processing, and automated mapping algorithms to detect, classify, and prioritize defects, eliminating the need for multiple human teams to manually analyze and resolve defects
Solution Approach 2:
The defect management engine performs self-service by automatically fetching defect data from repositories, mapping defects to features and test cases, calculating coverage metrics, and generating reports without requiring continuous human intervention, thereby reducing both detection incompleteness and resolution time
2Reliability
If comprehensive testing is performed to detect all defects, then defect detection completeness improves, but test case quantity and complexity increase significantly
Solution Approach 1:
The patent changes the parameter of test case management by using automated mapping algorithms that dynamically associate defects with relevant features and test cases based on learned patterns, rather than requiring manually created comprehensive test suites, thereby maintaining high detection completeness with reduced test case complexity
Solution Approach 2:
The defect management engine performs preliminary action by pre-fetching and analyzing defect data, feature information, and test case metadata before actual defect detection, using machine learning models to pre-classify and prioritize defects, which reduces the complexity of subsequent testing activities while maintaining comprehensive defect detection
Data Source
AI summary
Defect classification and association in a software development environment. The embodiments herein relates to a software development environment, and more particularly to managing defects in a software development environment. The defect management engine may use information present in the code repository to create a master list of features and/or sub-features. Using the master list, the defect management engine may scan and parse all reported defects (which may be present in the defect repository). The defect management engine may associate defects with features present in the master list. The defect management engine may then extract the test cases and map the test cases to the defects. The defect management engine may determine the test coverage metric and the test case sufficiency metric. The defect management engine may also label unmapped defects as orphan defects.