A Probabilistic Approach to Feature Selection for Multi-class Text Categorization.
Ke WuBao-Liang LuMasao UchiyamaHitoshi IsaharaPublished in: ISNN (1) (2007)
Keyphrases
- multi class
- text categorization
- feature selection
- text classification
- support vector machine
- information gain
- multi class classification
- naive bayes
- knn
- multi label
- mutual information
- automated text categorization
- k nearest neighbor
- feature weighting
- support vector
- probabilistic model
- bayesian networks
- binary classification
- generative model
- cost sensitive
- term frequency
- feature space
- binary classifiers
- multi task
- base classifiers
- feature extraction
- pairwise
- machine learning
- dimensionality reduction
- classification accuracy
- feature set
- feature selections
- class imbalance
- logistic regression
- document frequency
- unlabeled data
- class probabilities
- transfer learning
- nearest neighbor
- learning algorithm