Investigating Topic Models for Mobile Short Messaging Service Communication Filtering

This research work investigates the use of Latent Dirichlet Allocation (LDA), a generative topic modeling technique to extract latent features arising from mobile Short Messaging Service (SMS) communication for automatic discovery of user interest. This involves integrating temporal ordering of SMS into a generative process in an iterative manner. The mobile SMS documents are partitioned into segments, wherein the discovered topics in each segment are propagated to influence the discovery of latent features. The proposed technique filters malicious mobile SMS communication and shows that topic models can effectively detect distinctive latent features to support automatic content filtering overtime. The practical implication of SMS communication filtering is apparent in designing systems that proactively detect information security threats to mobile subscribers and operators. The system can assist in optimum decision making, for instance in a scenario where an imposture attempts to sneak confidential information from unsolicited messages send to a subscriber or an operator.