A novel dissimilarity metric based on feature-to-feature scatter frequencies for clustering-based feature selection in biomedical data

Abstract

Filter feature selection methods have been extensively used for dimensionality reduction in biomedical data analysis. In this article, a novel dissimilarity metric based on feature-to-feature (F2F) scatter frequencies is proposed for clustering-based filter feature selection. The proposed metric is computed by obtaining the feature-level ranks of samples and identifying the features which assign close ranks to each sample. The order of ranking is determined for each feature by class labels. Samples are represented as a set of affinity sets containing features having rank differences less than a predefined proximity window size. The F2F dissimilarity of a pair of features is computed using the frequency of their appearance in different affinity sets. Features are then clustered into distinct groups using F2F dissimilarity metric. From each cluster, the feature having the highest relevance score is selected. The experiments conducted on ten biomedical datasets confirmed the effectiveness of the proposed method in improving classification performance. Notably, the proposed method outperforms the widely used schemes in terms of both classification under ROC curve and stability.