Improvement of Fuzzy Geographically Weighted Clustering-Ant Colony Optimization using context-based clustering

Geo-demographic analysis (GDA) is an interdisciplinary that studies characteristics of population based on geographical area. Fuzzy Geographically Weighted Clustering-Ant Colony Optimization (FGWC-ACO), which is the improvement of FGWC algorithm, is considered as an effective algorithm in GDA. The integration of Ant Colony Optimization (ACO) as a metaheuristic algorithm to the FGWC has been used as a global optimization tools to improve geo-demographic clustering accuracy in initial phase. Nonetheless, using ACO makes the computation running time of FGWC-ACO is slower than standard FGWC. In this paper, we propose a method to attach context variables to FGWC-ACO in order to accelerate the computing speed of the algorithm and to focus the clustering result on a certain condition. An experiment of the proposed method has been done using Indonesia Population Census 2010 from Statistics Indonesia to prove that the proposed method can improve the computing speed of FGWC-ACO and using IFV index as a validity index for spatial fuzzy clustering to evaluate the clustering quality of proposed method.