COSDES of Junk E-Mail with Junk Free System Scheme

E-mail communication is indispensable now, but the e-mail spam problem is continuously growing more. In recent years, the notion of collaborative spam filtering with near-duplicate similarity matching scheme has been discussed widely. The idea of the similarity matching scheme for spam detection is, to maintain a database formed by user feedback and to block near-duplicate spams. The previous works mainly represent each e-mail by an abstraction derived from e-mail content text. These abstractions of emails cannot catch the evolving spams, and are thus not effective enough in near-duplicate detection. A procedure to generate the email abstraction using HTML content in e-mail, and newly devised abstraction which can be more efficient in capturing the duplicate phenomenon of spam is presented here. COSDES (COllaborative Spam DEtection System), a complete spam detection system, possesses an efficient near-duplicate matching scheme and a progressive update scheme. The forward-looking update scheme enables system COSDES to keep the most up-to-date information for near-duplicate detection. This system evaluates on a live data set collected from an e-mail server and shows that this system performs better than the previous approaches in detection results and is applicable to the real world.