基于一种改进的遗传算法的空间聚类分析

空间数据挖掘是数据挖掘的一个研究分支。空间聚类分析是空间数据挖掘的一个重要的研究领域。传统的K-均值方法用于聚类具有收敛速度快、算法实现简单等特点,但容易陷入局部最优,并对初始解敏感。遗传算法是一种全局搜索算法,但是收敛速度较慢。提出一种改进的遗传算法进行聚类,该算法通过全局搜索与局部搜索相结合,取得较好效果。实验表明:文中提出的算法在聚类分析中搜索到全局最优解(或近似全局最优解)的能力要优于经典的K-均值聚类算法,且局部收敛速度和全局收敛性能较好。

Spatial Clustering Analysis Based on an Improved Genetic Algorithm

Spatial data mining(SDM) is a branch of data mining(DM).Spatial clustering is an important field in SDM.Although the traditional K-means algorithm has good convergence rate and can be realized easily,it can easily be trapped in a local optimum,and it is sensitive in initial setting.Theoretically,genetic algorithm(GA) is a global optimization method,but with low convergence rate. Proposes an improved GA for spatial clustering.By integrating with global searching and local searching,it gets a better result.Experiments show that the ability of catch the global best solution(or approximate global best solution) of this algorithm in clustering analysis is better than classical K-means algorithm.And this new algorithm has better local convergence rate and global convergence performance.