Research Interest
My research interest is in data
mining and machine learning studying both the algorithmic and
application issues. The algorithmic aspects involve developing new
scalable, efficient and interactive algorithms that can handle very
large databases. The underlying techniques/tasks studied include
associations, classification, clustering, online optimization and
distributed learning. The application issues focus on actual
implementation and usage of the algorithms on a variety of data
sources with different characteristics including business
transactions, images, natural language, biology data, program trace
and data streams. My current work is focused on the following several
specific problems:
- Combining labeled and unlabeled data
- Multi-class (and Multi-label) classification approaches with applications in Text Categorization, Object Recognition and Bio-informatics
- Music information retrieval
- Association-based similarity and its applications
- Learning from multiple information sources