Spatial cluster analysis using particle swarm optimization and dispersion function.

Abstract

Spatial patterns studies are of great interest to the scientific community and the spatial scan statistic is a widely used technique to analyze such patterns. A key point for the construction of methods for detection of irregularly shaped clusters is that, as the geometrical shape has more degrees of freedom, some correction should be employed in order to compensate the increased flexibility. This paper proposed a multi-objective approach to cluster detection problem using the Particle Swarm Optimization technique aggregating a novel penalty function, called dispersion function, allowing only clusters which are subsets of a circular zone of moderate size. Compared to other regularity functions, the multi-objective scan with the dispersion function is faster and suited for the detection of moderately irregularly shaped clusters. An application is presented using statewide data for Chagas’ disease in puerperal women in Minas Gerais state, Brazil.