Optimizing Test Suites Using Ant Colony Optimization and Autoencoders for Efficient Defect Detection

Software testing plays a very significant role in providing software reliability, especially for large-scale distributed systems. Many test suite optimization techniques, like GA and RL, tend to incur high computational cost, take slow convergence rates, and incur redundancy in choosing the test cases. To overcome such challenges, this paper recommends an ACO and Autoencoder-based test suite optimization solution. The Autoencoder discovers the minimal test case characteristics without redundancy and preserving defect detection ability, while ACO selects an optimal subset of test cases based on defect detection efficiency. Hence, this enhances software testing effectiveness by prioritizing high-impact test cases while reducing execution time and improving fault detection. The experimental results show that the optimized test suite achieved an improvement of 23.2% in Defect Detection Efficiency (DDE), 26.1% in Fault Detection Rate (FDR), and 29.4% in Code Coverage (CC). The approach also reduces ET and Test Suite Size by 32.4% and 47.8% so that it could considerably boost testing execution efficiency.The results confirm that ACO and autoencoders are the best test case selectors; hence, they are highly suitable for large software test cases. The future could see hybrid optimization methods that combine the ACO with PSO or GA brought to bear in test case selection. Dynamic and adaptive test prioritization in multilevel software systems may use advanced deep learning techniques such as transformers and reinforcement learning. This paper sets a good ground for intelligent test suite optimization, enhancing defect detection and minimizing testing overhead.