ABSTRACT
Li-qiang Zhao, Shu-juan Wan, Zhi-guang Zhang, Huan Jiang, Xiao-rui Jia & Yan-hong Zhou School of Mathematics and Information Science & Technology, HeiBei Normal University of Science & Technology, Qinhuandao, Hebei Province, China
ABSTRACT: This article presents a biclustering method to deal with gene expression data based on Independent Component Analysis (ICA) and Non-negative Matrix Factorization (NMF). ICA is used to select genes to reduce the dimensionality of dataset and eliminate the influence from noisy or irrelevant genes, and NMF to deal with the chosen genes to achieve biclustering. The testing result on the gene expression datasets of Iris and Yeast indicates the eciency of this method.
