Detecting Cross-Site-Script Attacks using BM25 Algorithm

Abstract

Cross-Site-Script (XSS) attack is one of the main threats in the domain of web security. With the combination of artificial intelligence and web security, lots of the state-of-the-art methods have been proposed to detect XSS attacks. Following trends in attack detection methods, we proposed a detection method using BM25 algorithm, which can autonomously improve the efficiency of XSS detection. The key idea is to find an appropriate score in the calculation of the BM25 algorithm as the division standard for detection and judgment. The experiments show that our proposed method is feasible for XSS attack detection.