Title :
Improving clustering with hidden Markov models using Bayesian model selection
Author :
Li, C. ; Biswas, G.
Author_Institution :
Dept. of Electron. Eng. & Comput. Sci., Vanderbilt Univ., Nashville, TN, USA
Abstract :
This paper presents a Bayesian clustering methodology that partitions temporal data into homogeneous groups, and constructs state based profiles for each group in the hidden Markov model (HMM) framework. We propose a Bayesian HMM clustering methodology that improves upon existing HMM clustering algorithm by incorporating HMM model size selection into the clustering control structure. Experimental results indicate the effectiveness of our methodology
Keywords :
Bayes methods; hidden Markov models; pattern clustering; Bayesian clustering methodology; Bayesian model selection; hidden Markov model; homogeneous groups; state based profiles; temporal data partitioning; Automata; Bayesian methods; Casting; Character generation; Clustering algorithms; Data mining; Hidden Markov models; Partitioning algorithms; Size control; Stochastic processes;
Conference_Titel :
Systems, Man, and Cybernetics, 2000 IEEE International Conference on
Conference_Location :
Nashville, TN
Print_ISBN :
0-7803-6583-6
DOI :
10.1109/ICSMC.2000.884988
