Paper Info
Title
Minimum Classification Error Training for Online Handwritten Word Recognition
Abstract
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.
Year
DOI
Venue
2002
10.1109/IWFHR.2002.1030885
IWFHR
Keywords
Field
DocType
mce trainingachieves,writer variability,training criterion,minimum classification error training,word error rate reduction,unconstrained-style word recognition,minimum classification error,online handwritten word recognition,baseline maximum likelihood system,hidden markov models,probability,decision theory,signal processing,word recognition,handwriting recognition,speech recognition,writing,shape,word error rate,maximum likelihood,parameter estimation
Pattern recognition,Computer science,Word error rate,Word recognition,Maximum likelihood,Speech recognition,Decision theory,Artificial intelligence,Estimation theory,Hidden Markov model,Machine learning,Intelligent word recognition
Conference
ISBN
Citations 
PageRank 
0-7695-1692-0
9
0.77
References 
Authors
15
1
Name
Order
Citations
PageRank
Alain Biem128818.64