HMM Based Symbolic Sequence Self Organizing Clustering
基于隐马尔可夫模型的符号序列自组织聚类

In this paper, we propose a model-based, self organizing feature map algorithm for the clustering of variable-length sequences. Hidden Markov models(HMMs) are used as representations for the cluster centers, and batch map training algorithm is applied in clustering procedure. Simulation results show that our method can successfully find patterns of clusters of the input variable-length sequences.