一种基于图论的加权聚类融合算法

现有聚类融合算法对混合属性数据进行处理的效果不佳,主要是融合后的结果仍存在一定的分散性。为解决这个问题,提出了一种基于图论的加权聚类融合算法,通过对数据集聚类得到聚类成员后,利用所设计的融合函数对各个数据对象赋予权重,同时通过设置各个数据对间边的权重来确定数据之间的关系,得到带权最近邻图,再用图论的方法进行聚类。实验表明,该算法的聚类精度和稳定性优于其他聚类融合算法。

Weighted cluster fusion algorithm based on graph

The results of the existing cluster fusion algorithms are usually not so good when they process the mixed attributes datas, the main reason is that the results of the algorithms are still dispersed. To solve this problem, this paper presented a new weighted cluster fusion algorithm based on graph theory. It first clustered the datasets and got cluster members, and then set weights to each data object with a proposed fusion function, and determined the relationship between the data-pair by setting weights to the edges between them, so it could get a weighted nearest neighbor graph. At last it did a last-clustering based on graph theory. Experiments show that the accuracy and stability of this cluster fusion algorithm is better than other clustering fusion algorithms.