We show that a classifier based on Gaussian mixture models (GMM) can be trained discriminatively to improve accuracy. We describe a training procedure based on the extended Baum-W...
Class membership probability estimates are important for many applications of data mining in which classification outputs are combined with other sources of information for decisi...
Multi-view algorithms, such as co-training and co-EM, utilize unlabeled data when the available attributes can be split into independent and compatible subsets. Co-EM outperforms ...
Searching and organizing growing digital music collections requires a computational model of music similarity. This paper describes a system for performing flexible music similarit...
Michael I. Mandel, Graham E. Poliner, Daniel P. W....
An active set strategy is applied to the dual of a simple reformulation of the standard quadratic program of a linear support vector machine. This application generates a fast new...