Malicious VBScript Detection Algorithm Based on Data-Mining Techniques

Malware attacks are amongst the most common security threats. Not only malware incidents are rapidly increasing, but also the attack methodologies are getting more complicated. Moreover malware writers expand in using different platforms and languages. This raises the need for new detection methods which support more reliable, low resource consuming and fast solutions. In this paper, a new algorithm has been proposed based on machine learning techniques and static analysis features to detect malicious scripts specifically for VBScript files. Experimental results show that the proposed algorithm can achieve 97% detection ratio.