Named Entities as New Features for Czech Document Classification.
Pavel KrálPublished in: CICLing (2) (2014)
Keyphrases
- document classification
- named entities
- text documents
- text mining
- co occurrence
- named entity recognition
- information extraction
- named entity extraction
- natural language processing
- classification algorithm
- image features
- text classification
- question answering
- feature set
- text categorization
- web documents
- text analysis
- relation extraction
- maximum likelihood
- class labels
- model selection
- data sets
- classification accuracy
- pairwise
- training set
- feature extraction
- information retrieval