A Web Page Malicious Script Detection Method Inspired by the Process of Immunoglobulin Secretion

The immunoglobulin is one kind of special material for protecting the biological body far way from the invasion of harmful foreign antigen. When the biological body is under the attack by the continuous harmful foreign antigens, the set of immune cell in biological immune system will secret much immunoglobulin to neutralize them. Inspired by the process of immunoglobulin secretion in biological body, this paper presents a Web Page Malicious Script Detection Method inspired by the process of immunoglobulin secretion (WPMSD) to overcome the high false alarm rate in traditional immune-based web page malicious script detection methods. In WPMSD, Firstly, the definitions of immunoglobulin, antigen, self, nonself and detector in the web page malicious script detection domain are given. Secondly, according to the web page malicious script spreading range, the immunoglobulin number is changed from the detector clone proliferation under the continuous stimulations of web page malicious script. Thirdly, this paper proposes a probability approach for producing the web page malicious script alarm to notice the detected one is harmful. Finally, the WPMSD calculate the effective immunoglobulin set based on Hidden Markov Model (HMM) and update this effective immunoglobulin set into the detector gene library. Compared with traditional immune-based web page malicious script detection methods, such as Negative Selection Algorithm (NSA), Dynamic Clonal Selection (DynamiCS) and Variable size Detector (V-detector), the false alarm rate of WPMSD has been reduced.