 |
Improved estimation of hidden Markov model parameters from multiple observation sequences
Davis, Richard I. A., Lovell, Brian C. and Caelli, Terry (2002). Improved estimation of hidden Markov model parameters from multiple observation sequences. In: R. Kasturi, D. Laurendeau and C. Suen, Proceedings of the International Conference on Pattern Recognition. International Conference on Pattern Recognition, Quebec City, Canada, (168-171). 11-15 August, 2002.
|
|
| |
| Attached Files (Some files may be inaccessible until you login with your UQ eSpace credentials) |
| Name |
Description |
MIMEType |
Size |
Downloads |
n884.pdf
|
n884.pdf |
application/pdf |
51.66KB |
1949 |
| Author(s) |
Davis, Richard I. A.
Lovell, Brian C.
Caelli, Terry
|
| Title of paper |
Improved estimation of hidden Markov model parameters from multiple observation sequences
|
| Conference name |
International Conference on Pattern Recognition
|
| Conference location |
Quebec City, Canada
|
| Conference dates |
11-15 August, 2002
|
| Proceedings title |
Proceedings of the International Conference on Pattern Recognition
|
| Editor(s) |
R. Kasturi
D. Laurendeau
C. Suen
|
| Publisher |
The Institute of Electrical and Electronics Engineers
|
| Publication date |
2002
|
| Volume number |
2
|
| ISBN |
0-7695-1699-8
|
| Start page |
168
|
| End page |
171
|
| Total pages |
4
|
| Collection year |
2002
|
| Language |
eng
|
| Abstract/Summary |
The huge popularity of Hidden Markov models in pattern recognition is due to the ability to 'learn' model parameters from an observation sequence through Baum-Welch and other re-estimation procedures. In the case of HMM parameter estimation from an ensemble of observation sequences, rather than a single sequence, we require techniques for finding the parameters which maximize the likelihood of the estimated model given the entire set of observation sequences. The importance of this study is that HMMs with parameters estimated from multiple observations are shown to be many orders of magnitude more probable than HMM models learned from any single observation sequence - thus the effectiveness of HMM 'learning' is greatly enhanced. In this paper, we present techniques that usually find models significantly more likely than Rabiner's well-known method on both seen and unseen sequences.
|
| Subjects |
280200 Artificial Intelligence and Signal and Image Processing
|
| Keyword(s) |
Pattern recognition
Hidden Markov models
HMMs
iris-research
|
|
|
|
