Detection of malicious web pages based on hybrid analysis

Malicious web pages have become an increasingly serious threat to web security in recent years. In this paper, we propose a new detection method that consists of static and dynamic analyses for detecting malicious web pages. Static analysis utilizes classification algorithms in machine learning to identify certain benign and malicious web pages. As a complement to static analysis, dynamic analysis mainly checks the unknown web pages to determine whether they have malicious shellcodes during their execution. Because of the combination of static and dynamic analyses, the proposed detection method achieves high performance, and it has a light weight and is simple to use.