近邻传播算法(Affinity propagation,AP)是一种新提出的聚类算法,是在数据点的相似度矩阵的基础上进行聚类,通过数据点之间交换信息,最后得到聚类结果。
Affinity Propagation clustering algorithm (AP) is more flexible than K-means for three reasons.
First, AP does not require users to specify the number of clusters, when the number of aspects is just unknown.
Second, AP does not suff er from the initial selection of exemplars.
Third, AP can find representative sentences from S simultaneously.