Improvement of Fuzzy Geographically Weighted Clustering using Particle Swarm Optimization

One of the most interesting interdisciplinary research topics between Geographic Information Systems and Data Mining are known as Geo-demographic analysis (GDA). Among some methods in this field, Fuzzy Geographically Weighted Clustering (FGWC) is considered as the state-of-the-art for Geodemographic segmentation. In other hand, recent studies show that its quality of clustering achievement can be improved. One of its limitations is in cluster initialization phase, which performed randomly. In this paper, we introduce a hybrid approach of FGWC based on Particle Swarm Optimization (PSO), namely FGWC-PSO to overcome the limitation. The experimental evaluation through public data set studies shows that the proposed method obtains better clustering quality.