POSTDOC INTERVIEW TALK - Mining significant patterns on Dynamic Networks
- When:
- Monday, August 20, 2012, 11:00 am - 12:00 pm
- Where:
- 10th fl CR (1026)
- Speaker:
- Misael Mongiovi
- Description:
Host: Kristina Lerman
Title: Mining significant patterns on Dynamic NetworksBig network data abound in every area of our lives. Social networks, the Web, biological networks, are just a few of many examples of networks that impact our life. Analyzing the dynamics of such networks is a promising emergent research area. Dynamic networks are usually represented by a sequence of slices, or snapshots, where each slice embody the status of the network at a specific point in time. Mining dynamic networks is a challenging task since it often requires solving NP-hard problems on datasets with millions of nodes and edges.
One of the major challenges in this field is to extract meaningful patterns from dynamic networks efficiently. For instance, how can we identify spatio-temporal regions that are significant or anomalous in a dynamic network? Example of such regions are: set of users with increased rate of communication in a social network, unexpected congestions in road networks, contaminations in water networks. After a brief survey of the state of the art, I will present two novel algorithms for mining significant spatio-temporal regions on dynamic networks. These algorithms are able to efficiently handle large networks that evolve over long time periods. I will present some practical applications, with special emphasis on anomaly detection. I will conclude by introducing some ongoing work on mining smoothly evolving patterns.
Bio: Misael Mongiovi is a postdoctoral researcher at the Computer Science department at UC Santa Barbara, working on mining, management and optimizzation of large networks. Previously, he has been a postdoctoral scholar at Tel Aviv University and University of Catania, after receiving his Ph.D. in Computer Science. His main research interests concern Network mining, Graph querying, Social networks and Bioinformatics. Recently, his research work has focused mainly on mining significant patterns on dynamic networks.
