JOURNAL OF COMPUTERS (JCP)
ISSN : 1796-203X
Volume : 4    Issue : 11    Date : November 2009

A Research on Mixture Splitting for CHMM Based on DBC
Gang Liu, Wei Chen, and Jun Guo
Page(s): 1167-1174
Full Text:
PDF (417 KB)


Abstract
EM (expectation-maximization) algorithm is a classical method for parameter estimation of HMM
(Hidden Markov model). Concerning that EM algorithm is easily affected by initial parameter values,
a mixture splitting algorithm based on decision boundary confusion(DBC) was proposed to
describe more about  boundary distribution. The algorithm mainly includes four aspects:  firstly the
number of incremented mixtures for every decision boundary could be determined according to
decision boundary confusion; secondly the mixtures which are the closest to the decision boundary
are chosen to split; thirdly the split mean of mixture is in the direction of decision boundary; finally
the mixture number of a state is determined by the confusion between states. Our experiments
show that our proposed algorithm is more effective for classification using HMM.

Index Terms
mixture splitting, DBC, HMM, EM