An experimental evaluation of OCR text representations for learning document classifiers.
Markus JunkerRainer HochPublished in: Int. J. Document Anal. Recognit. (1998)
Keyphrases
- experimental evaluation
- learning algorithm
- document analysis
- decision trees
- learning process
- document processing
- document images
- printed documents
- text documents
- active learning
- information retrieval
- information extraction
- learning tasks
- feature selection
- feature representations
- support vector
- text categorization
- keywords
- training data
- optical character recognition
- machine learning