ScamSlam: An Architecture for Learning the Criminal Relations Behind Scam Spam

Provided by: Carnegie Mellon University
Topic: Software
Format: PDF
This white paper introduces Scam Slam, a software system designed to learn the underlying number criminal cells perpetrating a particular type of scam, as well as to identify which scam spam messages were written by which cell. The system consists of two main components; a filtering mechanism based on a Poisson classifier to separate scam from general spam and non-spam messages, and a message normalization and clustering technique to relate scam messages to one another. The paper involves application of Scam Slam to a corpus of approximately 500 scam messages communicating the advance fee fraud.

Find By Topic