Computer Science
8 years ago

Optimal classifier achieves highest recognition rates for classification tasks.

Communication in Statistics- Theory and Methods
A Bayes Empirical Bayes Decision Rule for Classification
Tze Fen Li, Tseng‐Chang Yen

Paper Summary

Related papers

Theme background image
Computer Science
2 years ago

Optimal Bayes Classifier Minimizes Error, Revolutionizing Pattern Recognition

Bayesian Networks
Data Mining Techniques
AI Planning
Theme background image
Computer Science
8 years ago

New method accurately predicts classification errors in pattern recognition technology.

Expert Systems
Theme background image
Computer Science
8 years ago

New method combines speed and accuracy for recognizing Chinese characters.

Learning with Noisy Labels
Dimensionality Reduction
Feature Retrieval

The researchers developed a new decision rule for classifying things that is as good as the best rule out there. They used a bunch of unknown samples to create this rule, and it works really well even when the number of unknown samples increases. Their rule gives recognition rates that are almost as good as the best rule that uses known parameters.