June 18, 2013 · 6:00 PM
CMU Professor Christos Faloutsos will review some of the techniques of anomaly detection and show how applying 'network effects'
to eBay transactions can be used to unearth hidden networks of eBay fraudsters.
Detecting anomalies and fraud has important applications in finance, online commerce, healthcare, law enforcement etc. Using network effects ('guilt by association') with large data sets is becoming increasingly effective in a range of analyses from identifying fraudulent transactions to uncovering social media players engaged in artificially boosting networks. Attend this session to see real world examples of graphical mining in action.