chapter  8
10 Pages

Variable Length Concentration-Based Feature Construction Method

In this chapter, we introduce the variable length concentrations (VLC)-based feature extraction method, which aims to break through the limit of feature vector’s length. Specifically, this method uses a fixed-length sliding window to divide each e-mail into several sections. The number of sections depends on the length of each e-mail. Consequently, the length of feature vectors varies from each other. This method can acquire adaptive feature vectors according to different lengths of e-mails.