Program : Dual Degree (Integrated Bachelors and Masters)
Advisor : Dr. Balaraman Ravindan
Areas of Interest : Machine Learning, Data Mining, Social Network Analysis
Research Topic : Transductive Learning using Multi-View Hypergraphs
Most traditional classification techniques, whether fully supervised or semi-supervised, treat data instances independently and train classifiers using their attribute values alone. But many a time real world data can be represented naturally in the form of networks and we often have access to multiple sets of attributes describing the same data, multiple relations between data instances and sometimes super-dyadic relations between them. This motivates us to use multi-view hypergraphs for collective classification in networks. Hypergraphs are a generalization of simple graphs and are a natural way of representing super-dyadic relations. The current focus of my work is to come up with a formalization for understanding the importance of different views and improving the classification performance.
Courses Completed: Machine Learning, Data Mining, Social Network Analysis, Natural Language Processing, Kernel Methods for Pattern Analysis