An Alternative Algorithm for Classification Large Categorical Dataset: K-Mode Clustering Reduced Support Vector Machine
The Reduced Support Vector Machine (RSVM) is extension method of Smooth Support Vector Machine (SSVM) for handling computational difficulties as well as reduces the model complexity by generating a nonlinear separating surface for a large dataset. To generate representative reduce set for RSVM, Clustering Reduced Support Vector Machine (CRSVM) was proposed. However, CRSVM is restricted to solve classification problems for large dataset with numeric attributes. In this paper, the authors propose an alternative algorithm, k-mode RSVM (KMO-RSVM) that combines RSVM and k-mode clustering technique to handle classification problems on categorical large dataset. Applying k-mode clustering algorithm to each class, they can generate cluster centroids of each class and use them to form the reduced set which is used in RSVM.