Mining Graphs and Time Series:
Patterns, Anomalies, and Fraud Detection

by Christos Faloutsos
CMU, Nov. 14, 2019
Tutorial to delegates from Government of India

Abstract

Given a large graph, like who-calls-whom,  or who-likes-whom,  what behavior is normal and what should be surprising, possibly due to fraudulent activity? How do graphs evolve over time? Given one or more time sequences, how can we forecast their evolution? how to spot anomalies and discontinuities?

We focus on these two topics: pattern and anomaly detection in static and time-evolving graphs; and analysis and forecasting in time sequences.

Foils