Suju Rajan
I am a Ph.D. student in the Department of Electrical and Computer Engineering at UT, Austin. I work with Dr.Joydeep Ghosh at the Laboratory for Artifical Neural Systems.

I am currently working in the following broad areas: design and application of multi-classifier systems, classifier ensembles and knowledge transfer, especially for hyperspectral data. My work is mostly done in collaboration with Dr. Melba Crawford's group at the Center for Space Research.

A Maximum Likelihood Framework for Integrating Taxonomies
Suju Rajan, Kunal Punera and Joydeep Ghosh, To appear in Proc. of the 20th National Conference on Artificial Intelligence, AAAI-05, Pittsburgh,USA.

Automatically Learning Document Taxonomies for Hierarchical Classification
Kunal Punera, Suju Rajan and Joydeep Ghosh, World Wide Web 2005, Chiba, Japan.

Exploiting Class Hierarchies for Knowledge Transfer in Hyperspectral Data
Suju Rajan and Joydeep Ghosh, In N.C. Oza, R. Polikar, J. Kittler and F. Roli editors, Multiple Classifier Systems, pages 417-428, LNCS Vol.3541, Springer 2005.

Empirical Evaluation of ECOC vs. BHC Approaches to Multiclass Problems
Suju Rajan, M.S. Thesis, University of Texas at Austin, May 2004.

Automatic Generation of Class Hierarchies for High-Dimensional Multiclass Problems
Joydeep Ghosh, Suju Rajan and Melba M. Crawford, Snowbird Learning Workshop, April 2004.

Empirical Evaluation of Hierarchical vs. Two-level Approaches to Multiclass Problems
Suju Rajan and Joydeep Ghosh, In F.Roli and J. Kittler editors, Multiple Classifier Systems, pages 283-292, LNCS Vol. 3077, Springer 2004.