摘要
为数据集选择合适的聚类算法是获得高质量聚类结果的前提和保障.提出了基于有效性指标的聚类算法选择方法,通过对不同聚类算法的聚类结果的质量评价为数据集选择最适合的聚类算法.该方法的优点是在对数据集的情况了解甚少的情况下,也能有效地保障聚类质量.实验结果表明本文方法十分有效,为实验数据集正确选择出最适合的聚类算法,并获得了高质量的聚类结果.
It is the precondition and guarantee for high clustering quality that are to select a proper clustering algorithm for a given dataset. The algorithm selection method based on a validity index is proposed. In this method, a proper algorithm is selected for a dataset through the quality evaluation of clustering results from different clustering algorithms. The proposed method has the advantage of guaranteeing clustering quality when there is less prior knowledge about the dataset. The experimental results show that the proposed method is effective and correctly finds out the proper clustering algorithm suiting for the experimental dataset.
出处
《四川师范大学学报（自然科学版）》
CAS
CSCD
北大核心
2011年第6期915-918,共4页
Journal of Sichuan Normal University（Natural Science）
基金
国家自然科学基金(60502047)
福建省教育厅基金(JA09043)资助项目
