Minimum Classification Error Training for Online Handwritten Word Recognition

I
IWFHR
2002
Eighth International Workshop on Frontiers in Handwriting Recognition (IWFHR'02)
Abstract - Minimum Classification Error Training for Online Handwritten Word Recognition

	This Article
	Subscribers, please Login Purchase article: $19 PDF IEEE Xplore Subscribers RSS feed
	Share
	Email this Article to a friend
	Bibliographic References
	ASCII Text BibTex RefWorks Procite/RefMan
	Add to:
	Digg Furl Spurl Blink Simpy Google Del.icio.us Y!MyWeb
	Search
	Similar Articles Articles by Alain Biem

Eighth International Workshop on Frontiers in Handwriting Recognition (IWFHR'02)

Ontario, Canada

August 06-August 08

ISBN: 0-7695-1692-0

	ASCII Text	x
	Alain Biem, "Minimum Classification Error Training for Online Handwritten Word Recognition," Frontiers in Handwriting Recognition, International Workshop on, pp. 61, Eighth International Workshop on Frontiers in Handwriting Recognition (IWFHR'02), 2002.

	BibTex	x
	@article{ 10.1109/IWFHR.2002.1030885, author = {Alain Biem}, title = {Minimum Classification Error Training for Online Handwritten Word Recognition}, journal ={Frontiers in Handwriting Recognition, International Workshop on}, volume = {0}, year = {2002}, isbn = {0-7695-1692-0}, pages = {61}, doi = {http://doi.ieeecomputersociety.org/10.1109/IWFHR.2002.1030885}, publisher = {IEEE Computer Society}, address = {Los Alamitos, CA, USA}, }

	RefWorks Procite/RefMan/Endnote	x
	TY - CONF JO - Frontiers in Handwriting Recognition, International Workshop on TI - Minimum Classification Error Training for Online Handwritten Word Recognition SN - 0-7695-1692-0 SP EP A1 - Alain Biem, PY - 2002 KW - null VL - 0 JA - Frontiers in Handwriting Recognition, International Workshop on ER -

Alain Biem, IBM T. J. Watson Research Center

DOI Bookmark: http://doi.ieeecomputersociety.org/10.1109/IWFHR.2002.1030885

We describe an application of the Minimum Classification Error (MCE) training criterion to online unconstrained-style word recognition. The described system uses allograph-HMMs to handle writer variability. The result, on vocabularies of 5k to 10k, shows that MCE training achieves around 17% word error rate reduction when compared to the baseline Maximum Likelihood system.

Citation:

Alain Biem, "Minimum Classification Error Training for Online Handwritten Word Recognition," iwfhr, pp.61, Eighth International Workshop on Frontiers in Handwriting Recognition (IWFHR'02), 2002

Peer Review Notice | Give Us Feedback

Usage of this product signifies your acceptance of the Terms of Use.