作者: Michael Schmid
DOI:
关键词: Associative property 、 Computer-aided 、 Data science 、 Cybercrime 、 Computer science 、 Modelling methods 、 The Internet 、 Authorship attribution 、 Attribution 、 World Wide Web 、 Writeprint
摘要: E-mail has become the most common way to communicate on Internet, but e-mail security and privacy mechanisms are still lacking. This proven be a very valuable characteristic for criminals, who can easily take advantage of e-mail’s various weaknesses remain anonymous. Consequently, cybercrime investigators need rely computer-aided writeprint modelling methods tools identify real author malicious e- mails with transformed semantic content. In this paper, we propose customized version associative classification, well-known data mining method, as well Support Count address authorship attribution problem. Experimental results real-life suggest that our proposed algorithms achieve good classification accuracy problem through use modelling.