Fast Feature Selection in an HMM-Based Multiple Classifier System for Handwriting Recognition.
Simon GünterHorst BunkePublished in: DAGM-Symposium (2003)
Keyphrases
- handwriting recognition
- multiple classifier systems
- feature selection
- classification accuracy
- machine learning
- writer independent
- hidden markov models
- speech recognition
- character recognition
- pattern recognition
- support vector machine
- text categorization
- word recognition
- classification systems
- word segmentation
- text classification
- feature set
- optical character recognition
- feature space
- base classifiers
- feature extraction
- majority voting
- learning algorithm
- naive bayes
- support vector
- texture classification
- bayesian networks
- face recognition
- handwritten words
- neural network
- machine vision
- knn