DYNAMIC TEST SELECTION AND PRIORITIZATION USING ASSOCIATION RULE MINING FOR CONTINUOUS INTEGRATION

Smart overview of the Invention

The patent application describes a system for dynamic test selection and prioritization using association rule mining. This system is designed to enhance the efficiency of software testing processes, particularly in the context of continuous integration. By analyzing a repository of test cases and their execution results, the system identifies relationships between test cases and uses this information to create optimized execution plans.

Process Description

The system begins by mining a test case repository to discover relationships based on historical execution data. It then formulates a first test case execution plan, selecting a subset of test cases deemed most critical. This plan is executed by a test management system, which runs the selected tests on one or more test systems. The results are analyzed to construct a second execution plan that aims to identify failures earlier in the testing process.

Field of Application

This method is particularly relevant for regression testing in complex information processing systems, such as those found in storage products. As software evolves, new features require validation without compromising existing functionality. The system addresses the challenge of limited testing resources by prioritizing the most valuable test cases to detect critical issues sooner.

Advantages Over Conventional Methods

Traditional testing methods often lack dynamic prioritization and fail to adapt to changes in software and testing contexts. They do not efficiently rank test cases for early failure detection or leverage adaptive association rule mining. The proposed system overcomes these limitations by dynamically selecting and prioritizing test cases based on specific software changes, thus improving testing coverage and efficiency.

Implementation Goals

The primary goals include ranking and prioritizing test cases for early failure detection, utilizing adaptive association rule mining techniques, and accommodating dynamic changes in software environments. These improvements aim to identify high-priority test cases that can uncover critical defects earlier in the testing cycle, reducing the risk of defects progressing to later stages.