Clustering Spatial Data Using a Kernel-Based Algorithm

Abstract

This paper presents a method for unsupervised partitioning of data using kernel methods which offer strength to deal with complex data non-linearly separable in input space. This work gets inspiration from the notion that a non-linear data transformation into some high dimensional feature space increases the possibility of linear separability of the patterns in the transformed space. Therefore, it simplifies exploration of the associated structure in the data. Kernel methods implicitly perform a non-linear mapping of the input data into a high dimensional feature space by replacing the inner products with an appropriate positive definite function. Firstly, in this paper, selective kernel-based clustering techniques are analyzed and their shortcomings are identified especially for spatial data analysis. Finally, we present a robust weighted kernel k-means algorithm incorporating spatial constraints for clustering spatial data as a case study. The proposed algorithm can effectively handle noise, outliers and auto-correlation in the spatial data. Therefore, this work comes up with new clustering algorithm using kernel-based methods for effective and efficient data analysis by exploring structures in the data.