log in  |  register  |  feedback?  |  help  |  web accessibility
Logo
Modeling complex network phenomena using discriminative network fragments
Prof. Ambuj Singh - University of Santa Barbara
Tuesday, June 11, 2013, 11:00 am-12:00 pm Calendar
  • You are subscribed to this talk through .
  • You are watching this talk through .
  • You are subscribed to this talk. (unsubscribe, watch)
  • You are watching this talk. (unwatch, subscribe)
  • You are not subscribed to this talk. (watch, subscribe)
Abstract

Global-state networks provide a powerful mechanism to model the increasing heterogeneity in data generated by current systems. Such a network comprises of a series of network snapshots with dynamic local states at nodes, and a global network state indicating the occurrence of an event. Mining discriminative subgraphs from global-state networks allows us to identify the in?uential subnetworks that have maximum impact on the global state and unearth the complex relationships between the local entities of a network and their collective behavior. We explore this problem and design a technique called MINDS to mine minimally discriminative subgraphs from large global-state networks. To combat the exponential subgraph search space, we propose the concept of an edit map and perform Metropolis Hastings sampling on the map to compute the answer set. Furthermore, we formulate the idea of network-constrained decision trees to learn prediction models on a subgraph while adhering to the underlying network structure. Extensive experiments on real datasets demonstrate excellent accuracy in terms of prediction quality. Additionally, MINDS achieves good speed-up over baseline techniques. An application area of this research is in understanding complex diseases where a set of pathways is perturbed. Understanding the logic of such perturbations from high-throughput datasets can help in elucidating the nature of diseases and their multiple states.

Bio

Ambuj Singh is a Professor of Computer Science and Biomolecular Science and Engineering at the University of California at Santa Barbara. He received a B.Tech. degree from the Indian Institute of Technology and a PhD degree from the University of Texas at Austin in 1989. His research interests are in querying and mining of large datasets, especially as they pertain to graphs, networks, high-dimensional and biological data. He has written over 180 technical papers in the areas of distributed computing, databases, and bioinformatics and graduated over 20 PhD students. He has led numerous interdisciplinary projects, and currently leads UCSB’s Information Networks Academic Research Center funded by the ARL.He has served on the editorial boards and program committees of several conferences, workshops and international meetings

This talk is organized by Amol