Using the absolute difference of term occurrence probabilities in binary text categorization.
Hakan AltinçayZafer ErenelPublished in: Appl. Intell. (2012)
Keyphrases
- text categorization
- absolute difference
- information gain
- occurrence probabilities
- term frequency
- document frequency
- term selection
- term weighting
- feature selection
- multi label
- text classification
- naive bayes
- k nearest neighbor
- text documents
- knn
- semi supervised learning
- distortion measure
- unlabeled data
- vector quantization
- document representation
- image classification
- probabilistic model
- pairwise
- data sets