Fusion of feature selection methods for pairwise scoring SVM

It has been recently discovered that stacking the pairwise comparison scores between an unknown patterns and a set of known patterns can result in feature vectors with nice discriminative properties for classification. However, such technique can be hampered by the curse of dimensionality because the vectors size is equal to the training set size. To overcome this problem, this paper investigates various filter and wrapper feature selection techniques for reducing the feature dimension of pairwise scoring matrices and argues that these two types of selection techniques are complementary to each other. A fusion technique is then proposed to combine the ranking criteria of filter and wrapper methods at algorithmic level. Evaluations on a subcellular localization benchmark demonstrate that feature sets selected by the fusion methods outperform those selected by the individual methods alone.