Density-Based Spatial Anomalous Window Discovery

The focus of this paper is to identify anomalous spatial windows using clustering-based methods. Spatial Anomalous are the contiguous groupings nodes which unusual with respect rest data. Many scan statistics based approaches have been proposed for identification windows. To similarly behaving groups points, clustering techniques proposed. There parallels between both types but these not used interchangeably. Thus, our work bridge gap and Specifically, we use circular statistic approach DBSCAN- Density Clustering Applications Noise, approach. We present experimental results in US crime data Our show that effective identifying performs equal or better than existing does pure clustering.