On the structure of hidden Markov models

被引:18
作者
Abou-Moustafa, KT
Cheriet, M
Suen, CY
机构
[1] Concordia Univ, Dept Comp Sci, CENPARMI, Montreal, PQ H3G 1M8, Canada
[2] Univ Quebec, Ecole Technol Super, LIVIA, Montreal, PQ H3C 1K3, Canada
基金
加拿大自然科学与工程研究理事会;
关键词
HMM structure; graphical models; credits diffusion; Ocham's razor; K-means clustering;
D O I
10.1016/j.patrec.2004.02.005
中图分类号
TP18 [人工智能理论];
学科分类号
081104 ; 0812 ; 0835 ; 1405 ;
摘要
This paper investigates the effect of HMM structure on the performance of HMM-based classifiers. The investigation is based on the framework of graphical models, the diffusion of credits of HMMs and empirical experiments. Although some researchers have focused on determining the number of states, this study shows that the topology has a stronger influence on increasing the performance of HMM-based classifiers than the number of states. (C) 2004 Elsevier B.V. All rights reserved.
引用
收藏
页码:923 / 931
页数:9
相关论文
共 34 条
[1]   Measuring HMM similarity with the Bayes probability of error and its application to online handwriting recognition [J].
Bahlmann, C ;
Burkhardt, H .
SIXTH INTERNATIONAL CONFERENCE ON DOCUMENT ANALYSIS AND RECOGNITION, PROCEEDINGS, 2001, :406-411
[2]  
BAKER JK, 1975, SPEECH SIGNAL PROCES, V23, P23
[3]  
BALASUBRAMANIAN V, 1993, 1370 MIT ART INT LAB
[4]   STATISTICAL INFERENCE FOR PROBABILISTIC FUNCTIONS OF FINITE STATE MARKOV CHAINS [J].
BAUM, LE ;
PETRIE, T .
ANNALS OF MATHEMATICAL STATISTICS, 1966, 37 (06) :1554-&
[5]  
BENGIO Y, 1999, NEURAL COMPUT SURVEY, V41, P129
[6]  
Bengio Y., 1995, NEURAL INFORMAT PROC, V7, P1251
[7]   Some new indexes of cluster validity [J].
Bezdek, JC ;
Pal, NR .
IEEE TRANSACTIONS ON SYSTEMS MAN AND CYBERNETICS PART B-CYBERNETICS, 1998, 28 (03) :301-315
[8]   A sequential pruning strategy for the selection of the number of states in hidden Markov models [J].
Bicego, M ;
Murino, V ;
Figueiredo, MAT .
PATTERN RECOGNITION LETTERS, 2003, 24 (9-10) :1395-1407
[9]  
Bicego M, 2001, LECT NOTES COMPUT SC, V2134, P75
[10]  
Biem A, 2003, PROC INT CONF DOC, P104