Weighted Features Classification with Pairwise Comparisons, Support Vector Machines and Feature Domain Overlapping

Most existing classification algorithms either consider all features as equally important, or do not analyze consistency of weights assigned to features. We show that by applying both Pairwise comparisons paradigm and Weighted Support Vector Machines, we can construct a classification algorithm where weights assigned to features are consistent. We start with pairwise comparisons to rank the importance of features, then we use distance-based inconsistency reduction to refine the weights assessment and make comparisons more precise. Finally, Weighted Support Vector Machines are used to classify the data. Also a new method of assigning weights to features, based on the concept of feature domain overlappings, is proposed and tested.