Survey on Existing Enterprise Web Application Security Mechanisms Using Machine Learning

Abstract

With the rapid growth of enterprise web applications, ensuring robust security has become a critical concern for organizations worldwide. Traditional security mechanisms struggle to keep up with the dynamic and evolving nature of web-based threats. Machine learning (ML) offers a promising approach by enabling intelligent and adaptive security solutions. This paper provides a comprehensive survey of existing enterprise web application security mechanisms that leverage ML techniques. It categorizes key ML-based security strategies, including intrusion detection, anomaly detection, fraud detection, and malware classification. Additionally, the survey highlights recent advancements, discusses performance evaluation metrics, and identifies current challenges such as data privacy, interpretability, and scalability. Future research directions are proposed to improve the efficiency and effectiveness of ML-driven web application security frameworks.