Graph Mining: Patterns and Tools
CMU, Tepper
April 4, 2017
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?
What tools are available, to find the most important nodes? the most strange nodes?
For the first, we present a list of static and temporal laws, and how they help us to find anomalies.
For the second, we show describe the powerful tool of Singular Value Decomposition (SVD), and we show how it can find important nodes (HITS, PageRank), as well as help with visualization and anomalies.
Foils
- Part 1: patterns
(in pdf and
in pps)
- Part 2: tools (SVD, etc)
(in pdf and
in pps )
Last updated by: Christos Faloutsos, April 2, 2017.