Fast Non-negative Dimensionality Reduction for Protein Fold Recognition

Abstract. In this paper, dimensionality reduction via matrix factorization with nonnegativity constraints is studied. Because of these constraints, it stands apart from other linear dimensionality reduction methods. Here we explore nonnegative matrix factorization in combination with a classifier for protein fold recognition. Since typically matrix factorization is iteratively done, convergence can be slow. To alleviate this problem, a significantly faster (more than 11 times) algorithm is proposed.

• Type: Conference
• Year: 2005
• Where: ECML
• Authors: Oleg Okun, Helen Priisalu, Alexessander Alves
• Keywords: Dimensionality Reduction, ECML 2005, Matrix Factorization, Nonnegative Matrix Factorization