CAJ | 학술논문

传统的k-means算法是一种局部搜索算法，对初始化敏感，容易陷入局部极值。针对此缺点，提出一种基于k-means算法的改进的蚁群聚类算法，选择相距最远的处于高密度区域的k个数据对象作为初始聚类中心，把正反馈、精英机制和变异算子引入到蚁群聚类。实验结果证明，算法不仅对初始数据具有弱依赖性，而且能够提高聚类的准确率，加快收敛。
전통적k-means산법시일충국부수색산법，대초시화민감，용역함입국부겁치。침대차결점，제출일충기우k-means산법적개진적의군취류산법，선택상거최원적처우고밀도구역적k개수거대상작위초시취류중심，파정반궤、정영궤제화변이산자인입도의군취류。실험결과증명，산법불부대초시수거구유약의뢰성，이차능구제고취류적준학솔，가쾌수렴。
The traditional k-means algorithm is a local search algorithm,which is sensitive to initializa-tion and easy to search a local maximum.To solve this problem,an improved ant colony clustering algorithm based on k-means algorithm is proposed.The algorithm selects k data objects which belong to high density area and are the furthest away from each other as initial center, and introduces positive feedback, elitism mechanism and mutation operator into ant colony clustering algorithm.Experiments show that the algorithm has not only the weak dependence on initial data,but also higher clustering accuracy and fast convergence.