中文摘要：

传统层次聚类算法中经常会遇到合并点和分裂点选择的问题,一旦一组对象被合并或者分裂,下一步的处理将在新生成类上进行,已做处理不能撤销,这样有可能导致低质量的聚类结果.针对这个问题,文中提出了一种模糊加权层次聚类改进算法,每次分层聚类时先计算对象属于这个类可靠度,然后和阀值进行比较,当可靠度小于阀值时重新确定对象的归属类,这样就解决了上述问题.最后通过实验验证,该算法确实可行有效.

英文摘要：

There is always the problem of choosing the merge point and split point in traditional hierarchical clustering algorithm.Once a group of objects are merged or split,the next step of the process will be done on the new category,in addition,the already done processing can＇t cancel,which could result in low quality clustering results.According to this problem,this paper proposes a weighted fuzzy hierarchical clustering algorithm.Every time when we stratify and cluster,we first calculate the reliability of the object to the category,and then compare it with the threshold.If the former is less than the latter,once again we determine its belonging category,which will solve the above problem.In the end,the experimental results show that this algorithm is feasible and effective.