A Cybercrime Forensic Method for Chinese Web Information Authorship Analysis

摘要

With the increasing popularization of the Internet, Internet services used as illegal purposes have become a serious problem. How to prevent these phenomena from happening has become a major concern for society. In this paper, a cybercrime forensic method for Chinese illegal web information authorship analysis was described. Various writing-style features including linguistic features and structural features were extracted. To classify the author of one web document, the SVM(support vector machine) algorithm was adopted to learn the author's features. Experiments on Chinese blog, BBS and e-mail data-set gained satisfactory results. The accuracy of blog dataset for seven authors was 89.49%. The satisfactory results showed that it was feasible to put the method to cybercrime forensic application.