Jove
Visualize
Contact Us
JoVE
x logofacebook logolinkedin logoyoutube logo
ABOUT JoVE
OverviewLeadershipBlogJoVE Help Center
AUTHORS
Publishing ProcessEditorial BoardScope & PoliciesPeer ReviewFAQSubmit
LIBRARIANS
TestimonialsSubscriptionsAccessResourcesLibrary Advisory BoardFAQ
RESEARCH
JoVE JournalMethods CollectionsJoVE Encyclopedia of ExperimentsArchive
EDUCATION
JoVE CoreJoVE BusinessJoVE Science EducationJoVE Lab ManualFaculty Resource CenterFaculty Site
Terms & Conditions of Use
Privacy Policy
Policies

Related Experiment Videos

Minimum classification error training for online handwriting recognition.

Alain Biem1

  • 1IBM T.J. Watson Research Center, PO Box 218, Yorktown Heights, NY 10598, USA. biem@us.ibm.com

IEEE Transactions on Pattern Analysis and Machine Intelligence
|June 24, 2006
PubMed
Summary
This summary is machine-generated.

Related Concept Videos

You might also read

Related Articles

Articles linked to this work by shared authors, journal, and citation graph.

Sort by
Same author

A discriminative approach to EEG seizure detection.

AMIA ... Annual Symposium proceedings. AMIA Symposium·2011
Same author

Semisupervised learning using Bayesian interpretation: application to LS-SVM.

IEEE transactions on neural networks·2011
Same author

Real-time prognosis of ICU physiological data streams.

Annual International Conference of the IEEE Engineering in Medicine and Biology Society. IEEE Engineering in Medicine and Biology Society. Annual International Conference·2010
Same author

Semisupervised least squares support vector machine.

IEEE transactions on neural networks·2009
Same author

Maximization of mutual information for offline Thai handwriting recognition.

IEEE transactions on pattern analysis and machine intelligence·2006
Same journal

Benchmarking the Robustness of Autonomous Driving to Environmental Illusions: A Lane Perception Perspective.

IEEE transactions on pattern analysis and machine intelligence·2026
Same journal

Learning Topology-Aware Representations via Test-Time Adaptation for Anomaly Segmentation.

IEEE transactions on pattern analysis and machine intelligence·2026
Same journal

TraGraph-GS: Trajectory Graph-based Gaussian Splatting for Arbitrary Large-Scale Scene Rendering.

IEEE transactions on pattern analysis and machine intelligence·2026
Same journal

SWIFT: A Small-World Interaction Framework for Flow-Aware Trajectory Prediction in Autonomous Driving.

IEEE transactions on pattern analysis and machine intelligence·2026
Same journal

HardFlow: Hard-Constrained Sampling for Flow-Matching Models Via Trajectory Optimization.

IEEE transactions on pattern analysis and machine intelligence·2026
Same journal

Industrial Brain: Self-Evolving Neuro-Symbolic Autonomy with Causal Resilience for Cyber-Physical Systems.

IEEE transactions on pattern analysis and machine intelligence·2026
See all related articles

This study applies Minimum Classification Error (MCE) training to online character and word recognition. MCE significantly reduces error rates compared to Maximum Likelihood methods, improving recognition accuracy for unconstrained writing styles.

Area of Science:

  • Computer Science
  • Artificial Intelligence
  • Pattern Recognition

Background:

  • Online character and word recognition is challenging due to unconstrained writing styles.
  • Traditional Maximum Likelihood (ML) methods may not optimally handle variations in handwriting.
  • The Minimum Classification Error (MCE) criterion offers a potential alternative for improving recognition accuracy.

Purpose of the Study:

  • To apply the Minimum Classification Error (MCE) criterion for online character and word recognition.
  • To evaluate the effectiveness of HMM-based MCE training in reducing error rates.
  • To assess the system's flexibility in handling diverse writing styles through multiple allographs.

Main Methods:

  • Implemented HMM-based character and word-level MCE training.

Related Experiment Videos

  • Utilized multiple allographs per character to accommodate writing style variations.
  • Conducted writer-independent experiments on alphanumeric characters and keyboard symbols.
  • Evaluated word recognition on vocabularies ranging from 5,000 to 10,000 words.

    • Main Results:

    • Achieved over 30% character error rate reduction compared to ML baseline.
    • Demonstrated approximately 17% word error rate reduction compared to ML baseline for large vocabularies.
    • Showcased improved recognition accuracy for unconstrained online handwriting.

    Conclusions:

    • MCE criterion is effective for improving online character and word recognition accuracy.
    • HMM-based MCE training offers significant error rate reduction over ML methods.
    • The approach enhances flexibility in recognizing diverse writing styles.