Title: A hybrid cross entropy method for spatial clustering problems
Authors: Nishanthi Raveendran - Macquarie University (Australia) [presenting]
Georgy Sofronov - Macquarie University (Australia)
Abstract: Spatial clustering is one of the important components of spatial data analysis. Spatial data are often heterogeneous, indicating that there may not be a unique simple statistical model describing the data. However, if we cluster the data into homogeneous clusters or domains, it will be easier to apply the appropriate statistical model for each domain. The problem of finding homogeneous domains is known as segmentation, partitioning or clustering. It is commonly used in many areas including disease surveillance, spatial epidemiology, population genetics, landscape ecology, crime analysis and many other fields. We focus on identifying homogeneous clusters and their boundaries in spatial data which is commonly used in epidemiological applications. To solve this clustering problem, we propose to combine the Cross-Entropy method, which is one of the evolutionary computing techniques that utilize a stochastic framework to solve estimation and a variety of optimization problems, with Voronoi tessellation to estimate the boundaries of such domains. Our results illustrate that the proposed algorithm is effective in identifying homogeneous clusters in spatial data.