A Spam Filtering Method Based on Multi-modal Features Fusion

In recent years, to escape the spam detection of the text-based spam filtering system, spammers insert junk information into the email with images, and attach it to the message body. The traditional text-based filter cannot handle such spam image. In order to deal with the spam which contains text and images, a filtering method which fuses text, image and other multi-modal features is proposed in this paper. Firstly, extracting the text features and image features to build multiple classifiers, and then by employing the fusion method to choose the output of multiple classifier. Experimental results on TREC dataset show that the fusion method can have a better result than that of a single classifier and can achieve over 90% in accuracy rate.