|
Research Area: Search Technologies, Machine Learning |
Profile
Deepayan Chakrabarti obtained his B.Tech. from IIT Kanpur, and his Ph.D. from Carnegie Mellon University. He has worked on graph mining problems, including epidemic thresholds in viral propagation, clustering and community detection on graphs, and building realistic graph generators.He is interested in applying such techniques on large graph datasets such as the Web and large social networks. Recently, he has also been working on bandit formulations for online learning.
Recent Publications, Projects and News
- Mortal Multi-Armed Bandits D. Chakrabarti; R. Kumar; F. Radlinski; E. Upfal, NIPS, 2008
- Generating Succinct Titles for Web URLs D. Chakrabarti; R. Kumar; K. Punera, KDD, 2008
- A Graph-Theoretic Approach to Webpage Segmentation D. Chakrabarti; R. Kumar; K. Punera, WWW, 2008
- Contextual Advertising by Combining Relevance with Click Feedback Deepayan Chakrabarti; Deepak Agarwal; Vanja Josifovski, WWW, 2008
- Page-level Template Detection via Isotonic Smoothing D. Chakrabarti; R. Kumar; K. Punera, WWW, 2007
- Estimating Rates of Rare Events at Multiple Resolutions D. Agarwal; A. Broder; D. Chakrabarti; D. Diklic; V. Josifovski; M. Sayyadian, KDD, 2007
- Page-Level Template Detection via Isotonic Smoothing Deepayan Chakrabarti; Ravi Kumar; Kunal Punera, 16th International World Wide Web Conference, 2007 [view abstract]
- Multi-armed Bandit Problems with Dependent Arms Pandey, S. ; Chakrabarti, D. ; Agarwal, D., ICML, 2007
- Bandits for Taxonomies: A Model-based Approach Pandey, S. ; Agarwal, D. ; Chakrabarti, D. ; Josifovski, V., SDM, 2007