A Novel K-Means Clustering Method for Locating Urban Hotspots Based on Hybrid Heuristic Initialization

With rapid economic and demographic growth, traffic conditions in medium large cities are becoming extremely congested. Numerous metropolitan management organizations hope to promote the coordination of urban development by formulating improving strategies. The effectiveness these solutions depends largely on an accurate assessment distribution hotspots (centers activity). In recent years, many scholars have employed K-Means clustering technique identify hotspots, believing it be efficient. K-means is a sort iterative analysis. When data dimensionality sample size enormous, algorithm sensitive initial centers. To mitigate problem, hybrid heuristic “fuzzy system-particle swarm-genetic” algorithm, named FPSO-GAK, obtain better centers for algorithm. results evaluated analyzed using three-cluster evaluation indexes (SC, SP SSE) two-cluster similarity (CI CSI). A taxi GPS dataset multi-source were test validate proposed comparison Random Swap (RS), Genetic (GAK), Particle Swarm Optimization (PSO) based K-Means, PSO constraint Weighted PSO-GA K-Means++ algorithms. findings demonstrate that can achieve results, as well successfully acquire hotspots.