Spatial Clustering with Obstacles Constraints by HPSO based on Grid

Spatial clustering has been an active research area in the data mining community. Spatial clustering is not only an important effective method but also a prelude of other task for spatial data mining (SDM).In this paper, we propose a novel spatial clustering with obstacles constraints (SCOC) using an advanced hybrid particle swarm optimization (HPSO) with GA mutation based on grid model. In the process of doing so, we first developed a novel spatial obstructed distance using HPSO based on grid model (HGSOD) to obtain obstructed distance, and then we presented a new HPKSCOC based on HPSO and K-Medoids to cluster spatial data with obstacles constraints. The experimental results show that HGSOD is effective, and HPKSCOC can not only give attention to higher local constringency speed and stronger global optimum search, but also get down to the obstacles constraints and practicalities of spatial clustering; and it performs better than improved K-Medoids SCOC (IKSCOC) in terms of quantization error and has higher constringency speed than genetic K-Medoids SCOC (GKSCOC).