Leveraging Propagation for Data Mining: Models, Algorithms and Applications

Naren Ramakrishnan

Abstract

Can we infer if a user is sick from her tweet? How do opinions get formed in online forums? Which people should we immunize to prevent an epidemic as fast as possible? How do we quickly zoom out of a graph? Graphs---also known as networks---are powerful tools for modeling processes and situations of interest in real life domains of social systems, cyber-security, epidemiology, and biology. They are ubiquitous, from online social networks, gene-regulatory networks, to router graphs.

This tutorial will cover recent and state-of-the-art research on how propagation-like processes can help big-data mining specifically involving large networks and time-series, algorithms behind network problems, and their practical applications in various diverse settings. Topics include diffusion and virus propagation in networks, anomaly and outbreak detection, event prediction and connections with work in public health, the web and online media, social sciences, humanities, and cyber-security.