SCS Undergraduate Thesis Topics

Student Advisor Thesis Topic
Tom Shen William Cohen Graph-Based Semi-Supervised Learning for Text Categorization through Supervised Random Walks

Recently, many effective graph-based semi-supervised learning methods have been developed. For text categorization, a common method is label propagation through a bipartite graph of documents and features or a k-nearest neighbor graph of documents. In this talk, we consider a new approach based on supervised random walks.

Close this window