摘要
在聚类分析中,模糊k 均值算法是目前应用最为广泛的方法之一,然而该算法对初始化敏感,容易陷入局部极值点.为此,提出一种基于克隆选择的模糊聚类新算法以实现全局优化处理.在新算法中,由于克隆算子能够将进化搜索与随机搜索、全局搜索和局部搜索相结合,因而通过对候选解进行克隆算子操作,能够快速得到全局最优解.用人造数据和IRIS实际数据所做测试结果表明了新算法的有效性.
In cluster analysis, fuzzy k-means (FKM) algorithm is one of the most widely used methods. However, FKM algorithm is much more sensitive to the initialization, and easy to fall into local optimum. For this purpose, it presents a clonal selection based new algorithm for fuzzy clustering analysis, for global optimization. Since the clonal operator can combine the evolutionary search and random search, and incorporate the global search with local search, by the clonal operation on candidate solutions, the new algorithm can quickly obtain the global optimum. The experimental results with synthetic data and IRIS real data illustrate the effectiveness of the new algorithm.
出处
《复旦学报（自然科学版）》
CAS
CSCD
北大核心
2004年第5期815-818,共4页
Journal of Fudan University：Natural Science
基金
ProjectsupportedbyNSFC(6 0 2 0 2 0 0 4 )
