---

* News

Colloquia

Perspective on Link Prediction

Nitesh Chawla
Notre-Dame University

February 16, 2011
DARRIN 337- 2:00 p.m. to 3:00 p.m.

Abstract:


Link prediction is the task of predicting relationships in a network. As interest in network science grew, so did the realization of the broad applicability of general link prediction. Link prediction has indeed found a remarkable array of applications in network science --- from security to collaboration to marketing to information flow to biology and medicine. Such broad applicability also brings forth a number of challenges to consider, including generality of methodologies, modes of evaluation, and scalability. In this talk, I offer a perspective on link prediction, providing a set of best practices for network construction, testing, and evaluation. I will also present a number of different prediction models, including our own supervised link predicted on a large number of networks. These networks range from a size of 400 nodes to 7.8 million nodes, and collectively represent a number of different applications of link prediction.

Bio:


Dr. Nitesh Chawla is an Assistant Professor in the Department of Computer Science and Engineering at the University of Notre Dame. He directs the Data Inference Analysis and Learning Lab (DIAL) and co-directs the Interdisciplinary Center for Network Science and Applications (iCeNSA) at Notre Dame. His research is primarily focused on data mining, machine learning and network science, and the inter-disciplinary applications of the same, including climate data sciences, social networks, and healthcare informatics. He is a recipient of number awards and acknowledgments, including outstanding dissertation award, NIPS classification challenge, best paper awards, the NAE New Faculty Fellowship, and outstanding undergraduate teacher award. His research has been supported from NSF, DOD, ARL, NIJ, and industry sponsors.

Hosted by: Dr. Sibel Adali (x8047)

Last updated: February 12, 2011


---

---