On enhancing the performance of spam mail filtering system using semantic enrichment

With the explosive growth of the Internet, e-mails are regarded as one of the most important methods to send e-mails as a substitute for traditional communications. As e-mail has become a major mean of communication in the Internet age, exponentially growing spam mails have been raised as a main problem. As a result of this problem, researchers have suggested many methodologies to solve it. Especially, Bayesian classifier-based systems show high performances to filter spam mail and many commercial products available. However, they have several problems. First, it has a cold start problem, that is, training phase has to be done before execution of the system. The system must be trained about spam and non-spam mail. Second, its cost for filtering spam mail is higher than rule-based systems. Last problem, we focus on, is that the filtering performance is decreased when E-mail has only a few terms which represent its contents. To solve this problem, we suggest spam mail filtering system using concept indexing and Semantic Enrichment. For the performance evaluation, we compare our experimental results with those of Bayesian classifier which is widely used in spam mail filtering. The experimental result shows that the proposed system has improved performance in comparison with Bayesian classifier respectively.