 |
Improved Classification Using Hidden Markov Averaging From Multiple Observation Sequences
Davis, R. I. A., Walder, C. J. and Lovell, Brian C. (2002). Improved Classification Using Hidden Markov Averaging From Multiple Observation Sequences. In: V. Chandran, Proceedings of the Fourth Australasian Workshop on Signal Processing and Applications 2002. Fourth Australasian Workshop on Signal Processing and Applications 2002, Brisbane, (89-92). 17-18 December, 2002.
|
|
| |
| Attached Files (Some files may be inaccessible until you login with your UQ eSpace credentials) |
| Name |
Description |
MIMEType |
Size |
Downloads |
richarddaviswosp.pdf
|
richarddaviswosp.pdf |
application/pdf |
66.04KB |
655 |
| Author(s) |
Davis, R. I. A.
Walder, C. J.
Lovell, Brian C.
|
| Title of paper |
Improved Classification Using Hidden Markov Averaging From Multiple Observation Sequences
|
| Conference name |
Fourth Australasian Workshop on Signal Processing and Applications 2002
|
| Conference location |
Brisbane
|
| Conference dates |
17-18 December, 2002
|
| Proceedings title |
Proceedings of the Fourth Australasian Workshop on Signal Processing and Applications 2002
|
| Editor(s) |
V. Chandran
|
| Place published |
Brisbane
|
| Publisher |
Queensland University of Technology
|
| Publication date |
2002
|
| Volume number |
4
|
| ISBN |
1 74107 002 3
|
| Start page |
89
|
| End page |
92
|
| Total pages |
4
|
| Collection year |
2002
|
| Language |
eng
|
| Abstract/Summary |
The enormous popularity of Hidden Markov models (HMMs) in spatio-temporal pattern recognition is largely due to the ability to 'learn' model parameters from observation sequences through the Baum-Welch and other re-estimation procedures. In this study, HMM parameters are estimated from an ensemble of models trained on individual observation sequences. The proposed methods are shown to provide superior classification performance to competing methods.
|
| Subjects |
280200 Artificial Intelligence and Signal and Image Processing
E1
290901 Electrical Engineering
780101 Mathematical sciences
|
| Keyword(s) |
HMM
Hidden Markov models
iris-research
|
|
|
|
