A new hierarchical cluster validity index based on IB method
Lingling Niu, Zhengzheng Lou, Yongdong Ye · 2010
To solve the problem of determining the correct number of clusters, this paper proposes a new cluster validity index, IB_Hindex, for hierarchical clustering based on IB method. The index effectively incorporates the cluster cohesion and separation so that the corresponding algorithm is able to find the number of feature patterns hidden in dataset. IB_Hindex is applied to binary hierarchical clustering algorithm. We observe that the algorithm can detect the reasonable number of clusters in the uniform and nonuniform data, even without any input parameters.