A Review of Machine Learning Approaches for Software Testing in Secure Modern Systems

Abstract

Software testing plays a critical role in ensuring the quality, reliability, and security of modern software systems. With the increasing complexity of applications driven by cloud computing, the Internet of Things (IoT), mobile technologies, and distributed architectures, traditional testing approaches often face challenges related to scalability, efficiency, and cost. Machine Learning (ML) has emerged as a promising solution to enhance software testing by automating testing activities and improving decision-making processes. This review paper presents a comprehensive analysis of ML-based approaches in software testing for secure modern systems. It examines key applications of ML, including defect prediction, fault detection, automated test case generation, test optimization, and vulnerability assessment. The paper also discusses the integration of Artificial Intelligence (AI) and Large Language Models (LLMs) in software testing and quality assurance. Furthermore, it explores security testing methodologies and emerging trends such as DevSecOps and continuous security testing. The review highlights the benefits, challenges, and limitations of current approaches while identifying future research directions for developing intelligent, scalable, and security-aware software testing frameworks that improve software quality and testing efficiency.