摘要
将自适应蚁群优化算法与FCM(Fuzzy C-Means)算法相结合,提出了一种模糊聚类分析的新算法.该算法通过把FCM算法中的目标函数降维,将其转化为自适应蚁群优化算法中的优化函数,通过对各个节点的路径连接数的衡量,根据蚂蚁在搜索过程中所得解的分布状况,动态调节蚂蚁的路径选择和信息量更新,从而得到目标函数的最优解.结果表明,该方法比FCM算法具有更好的收敛效果和更高的聚类准确率.
Combining self-adaptive ant colony optimization algorithm with the FCM algorithm,a new fuzzy clustering algorithm is proposed.The dimensions of the objective function in the FCM algorithm are reduced by this method to convert into self-adaptive ant colony optimization algorithm optimization function.Furthermore,according to the solutions of the distribution,which are obtained in the search process by ants,the path is chosen by ants and the updated amount of pheromone is adjusted dynamically.Thus,the optimal solutions of the objective function are obtained.The results show that this method has the better convergence effect and the higher clustering accuracy than the FCM algorithm.
出处
《华北水利水电学院学报》
2011年第6期134-137,共4页
North China Institute of Water Conservancy and Hydroelectric Power
基金
国家自然科学基金项目(10761008)
关键词
蚁群算法
模糊聚类
连续空间优化
FCM
信息素
正反馈
ant colony algorithm
fuzzy clustering
continuous space optimization
FCM
pheromone
positive feedback