Joint Clustering and Feature Selection

Due to the absence of class labels, unsupervised feature selection is much more difficult than supervised feature selection. Traditional unsupervised feature selection algorithms usually select features to preserve the structure of the data set. Inspired from the recent developments on discriminative clustering, we propose in this paper a novel unsupervised feature selection approach via Joint Clustering and Feature Selection (JCFS). Specifically, we integrate Fisher score into the clustering framework. We select those features such that the fisher criterion is maximized and the manifold structure can be best preserved simultaneously. We also discover the connection between JCFS and other clustering and feature selection methods, such as discriminative K-means, JELSR and DCS. Experimental results on real world data sets demonstrated the effectiveness of the proposed algorithm.