Differential Voting in Case Based Spam Filtering

Case-based reasoning (CBR) has been shown to be of considerable utility in a spam-filtering task. In the course of this study, we propose that the non-random skewed distribution of the cases in a case base is crucial, especially in the context of a classification task like spam filtering. In this paper, we propose approaches to improve the performance of a CBR spam filter by making use of the non-random nature of the case base. We associate each case in the case base with a voting power, which is essentially a function that incorporates the knowledge of the local neighborhood of the case. We show that the performance of the spam filter can be considerably improved by making use of such techniques that incorporate the voting powers.