Efficient training of direct multi-class formulations of linear Support Vector Machines is very useful in applications such as text classification with a huge number examples as w...
S. Sathiya Keerthi, S. Sundararajan, Kai-Wei Chang...
Censored targets, such as the time to events in survival analysis, can generally be represented by intervals on the real line. In this paper, we propose a novel support vector tec...
Pannagadatta K. Shivaswamy, Wei Chu, Martin Jansch...
This paper presents an efficient hybrid feature selection model based on Support Vector Machine (SVM) and Genetic Algorithm (GA) for large healthcare databases. Even though SVM an...
Rick Chow, Wei Zhong, Michael Blackmon, Richard St...
Natural Language Processing (NLP) for Information Retrieval has always been an interesting and challenging research area. Despite the high expectations, most of the results indica...
Kernel machines rely on an implicit mapping of the data such that non-linear classification in the original space corresponds to linear classification in the new space. As kernel ...