An AI-ML-powered high quality engineering method makes use of AI-ML to boost software program high quality assessments by predicting defects. Present ML fashions wrestle with noisy information sorts, imbalances, sample recognition, characteristic extraction, and generalization. To deal with these challenges, we develop a brand new mannequin, Adaptive Differential Evolution (ADE) primarily based Quantum Variational Autoencoder-Transformer (QVAET) Mannequin (ADE-QVAET). ADE combines with QVAET to acquire high-dimensional latent options and keep sequential dependencies, leading to enhanced defect prediction accuracy. ADE optimization enhances mannequin convergence and predictive efficiency. ADE-QVAET integrates AI-ML strategies corresponding to tuning hyperparameters for scalable and correct software program defect prediction, representing an AI-ML-driven know-how for high quality engineering. Throughout coaching with a 90% coaching share, ADE-QVAET achieves excessive accuracy, precision, recall, and F1-score of 98.08%, 92.45%, 94.67%, and 98.12%, respectively, when in comparison with the Differential Evolution (DE) ML mannequin.
