欢迎您！东篱公司
退出
-
申报数据库
-
立项数据库
-
成果数据库
-
项目获奖数据库
中文摘要：
为了进一步提升ESSC聚类融合性能，采用实数值链接分析（real valued link analysis）计算聚类融合中模糊数据类的相似性。根据模糊决策及其相似性定义优化的融合信息，从而达到改进聚类性能的目的。实验选用了两个仿真数据库和五个UCI数据库。实验结果表明，基于实数值链接分析的ESSC聚类融合算法（RLA-ESSCE）的性能优于K-means聚类算法（KMC）、ESSC、ESSCE。
英文摘要：
In order to further improve the performance of ESSCE, real-valued link analysis had been proposed to compute the similarity between fuzzy clusters in ESSC clustering ensemble （RLA-ESSCE）. The clustering ensemble information was refined according to fuzzy decision and its similarity. Therefore the performance according to refined clustering ensemble information had been improved. Experiments were conducted on two synthetic datasets and five UCI datasets. Experimental results show that RLA-ESSCE is better than K-means clustering （KMC） ,ESSC and ESSCE.
同期刊论文项目
同项目期刊论文
期刊信息
位置：