An unsupervised feature ranking scheme by discovering biclusters

In this paper, we aim to propose an unsupervised feature ranking algorithm for evaluating features using discovered biclusters which are local patterns extracted from a data matrix. The biclusters can be expressed as sub-matrices which are used for scoring relevant features from two aspects, i.e. the interdependence of features and the separability of instances. The features are thereby ranked with respect to their accumulated scores from the total discovered biclusters before the pattern classification. Experimental results show that this proposed algorithm can yield comparable or even better performance in comparison with the well-known Fisher Score, Laplacian Score and Variance Score using several UCI data sets.