Short Messages Spam Filtering Using Personality Recognition

Abstract

In this era of smartphones and online social networks, short messages communications tools are growing up, attracting spam and non-legitimate campaigns. Those campaigns, besides being an illegal online activities, are a direct threat to the privacy of the users. Previous short messages spam filtering techniques focus on automatic text classification and while none of them take message personality feature into account. Focusing on phone Short Message Service (SMS) messages, this work demonstrates that it is possible to improve spam filtering in short message services using personality recognition techniques. Using a publicly available labelled (spam/legitimate) SMS dataset, we apply personality recognition techniques to each message and aggregate the personality feature to the original dataset, creating a new one. We compare the results of the best classifiers and filters over the different datasets (with and without personality) in order to demonstrate the influence of the personality. Experiments show that personality feature helps to improve SMS spam filtering improving both the accuracy, reaching to a 98.94%, and the false positive rates.