Selecting subspace dimensions for kernel-based nonlinear subspace classifiers using intelligent search methods

In Kernel based Nonlinear Subspace (KNS) methods, the subspace dimensions have a strong influence on the performance of the subspace classifier. In this paper, we propose a new method of systematically and efficiently selecting optimal, or near-optimal subspace dimensions for KNS classifiers using a search strategy and a heuristic function termed as the Overlapping criterion. The task of selecting optimal subspace dimensions is equivalent to finding the best ones from a given problem-domain solution space. We thus employ the Overlapping criterion of the subspaces as a heuristic function, by which the search space can be pruned to find the best solution to reduce the computation significantly. Our experimental results demonstrate that the proposed mechanism selects the dimensions efficiently without sacrificing the classification accuracy. The results especially demonstrate that the computational advantage for large data sets is significant.