Abstract:
Detection of spam emails is a key task since the internet community highly suffers from spam emails as nearly 90% of the incoming emails are spam. In this paper, a Cooperative Vector-based Reactive System (CVRS) has been proposed which filters spam emails in three steps, email classification, similarity detection and cooperative reaction. The CVRS system has been implemented at the receiver side with a group of reporters that evaluate the reactive feedback send by those reporters in a cooperative fashion. The CVRS model accurately filters spam emails without any delay. The CVRS system has been implemented using Map Reduce functionality and its performance has been evaluated using metrics such as false positive rate, false negative rate, detection accuracy and detection time. CVRS calculates feature probability on the clustered email, hence this creates only a short detection delay. Furthermore, CVRS system reduces the number of false positives and negatives by calculating similarity detection on the clustered email and thus achieves a high accuracy through validating the reporter’s feedback result.
