Date: Friday, January 29, 2010
Reception: 3:00 – 3:30pm Duthie Center 2nd Floor
Seminar 3:30, Duthie Room 117

Non-negative Matrix Factorization for Multimodal Image Representation

Abstract:
Visual content has become an important component of the web. In many cases, visual content is mixed with other modalities (e.g. text) that can be exploited to extract information and knowledge. This talk presents a strategy that uses Non-Negative Matrix Factorization (NMF) to generate a rich representation of multimodal objects. NMF has been used to address different problems in computer vision, machine learning and text-mining, among others. Recently however, it has received even more considerable attention thanks to its successful application to building a recommender system based on collaborative filtering that finally won the NetFlix price. The talk will discuss the basics of NMF, how it relates to other clustering and dimensionality reduction algorithms, and how it could be applied to the problem of multimodal image retrieval.

Bio:
Fabio A. Gonzalez is an Associate Professor at the Department of Computer Systems and Industrial Engineering, National University of Colombia. He is the co-leader of the Bioingenium research group. He earned a Computer Systems Engineer degree and a MSc in Math degree from the National University of Colombia in 1993 and 1998 respectively, and a MSc and PhD degrees in Computer Science from the University of Memphis, USA, in 2003. His research work is mainly focused on the foundations of machine learning and its applications to image processing, computer vision, data mining and information retrieval among others.