Clustering spatial data using random walks
Proceedings of the seventh ACM SIGKDD international conference on Knowledge discovery and data mining
On Clustering Using Random Walks
FST TCS '01 Proceedings of the 21st Conference on Foundations of Software Technology and Theoretical Computer Science
Hi-index | 0.00 |
Discovering significant patterns that exist implicitly in huge spatial databases is an important computational task. A common approach to this problem is to use cluster analysis. However, traditional clustering methods have several shortcomings when addressing spatial data. We propose a novel approach to clustering spatial data, based on the deterministic analysis of random walks on a weighted graph generated from the data. Our approach has several advantages. First, it can decompose the data into arbitrarily shaped clusters of different sizes and densities. Second, it can overcome noise and outliers that may blur the natural decomposition of the data. Third, our method requires only O(n*log n) time, and one of its variants needs only constant space.