DMEA-II and its application on spam email detection problems
Long Nguyen, Anh Quang Tran, Lam Thu Bui · 2014
This paper deals with multi-objectivitv in the problem of Vietnamese spam detection. We first analyze the problem taking into account the specific Vietnamese characterises as well as multi-objectivity, With the use of multi-objectivity, we can allow the users more flexibility on selecting the solution. Our proposal is to extende a multi-objective optimization algorithm using directional information, called DMEA-II for finding sets of feasible trade-off solutions for an anti-spam email system (using Apache SpamAssassin). The two objectives for considering are the Spam Detection Rate (SDR) and False Alarm Rate (FAR). The experiments were conducted based on spam data sets through several scenarios with different numbers of SpamAssassin rules. According to the obtained results, the new approach based on DMEA-II not only achieved more efficient results but also created a set of ready-to-use rule scores trading-off between SDR and FAR. It demonstrates the ability to give users more flexibility and efficiency in the Anti-spam email System.