Abstract: Clustering algorithms have many useful applications in signal processing and pattern recognition. In conventional clustering methods, we classify objects in terms of all features available and measure the similarity between two objects using a distance metric. In biclustering, we detect coherent patterns in both object and feature directions and we are interested in data consistency, such as simultaneous downward or upward changes of a subset of features for a subset of objects, rather than the distances among all objects for all features. Biclustering is naturally more difficult than clustering computationally and is often considered intractable mathematically. We have recently developed hyperplane based methods for the detection of a class of biclusters in a high-dimensional signal space. Our methods provide a unified model for several types of biclusters and can be implemented using efficient signal processing algorithms. We have also found an interesting link between biclustering and the spectral graph theory. We have applied our biclustering methods to disease diagnosis and drug therapeutic effect assessment using DNA microarray gene expression data. For example, we are able to identify biclusters of a subset of genes that co-express under a subset of conditions. These biclusters are useful for the identification of caner types and sub-types. We have also implemented biclustering algorithms on the field-programmable gate array (FPGA) for fast computation. In this seminar, the recent work of our research group on biclustering methods and their applications will be presented.

Brief Biography of the Speaker:
Hong Yan received his Ph.D. degree from Yale University. He was Professor of Imaging Science at the University of Sydney and is currently Professor of Computer Engineering at City University of Hong Kong. His research interests include image processing, pattern recognition and bioinformatics, and he has over 300 journal and conference publications in these areas. Professor Yan was elected an IAPR fellow for contributions to document image analysis and an IEEE fellow for contributions to image recognition techniques and applications.