Weighted Average Pointwise Mutual Information for Feature Selection in Text Categorization.
Karl-Michael SchneiderPublished in: PKDD (2005)
Keyphrases
- text categorization
- weighted average
- pointwise mutual information
- feature selection
- automated text categorization
- text classification
- multi label
- knn
- weighted sum
- information gain
- mutual information
- text filtering
- text documents
- feature weighting
- reuters corpus
- naive bayes
- tf idf
- feature reduction
- feature generation
- automatic text categorization
- term weighting
- feature extraction
- feature selections
- text classifiers
- neural network
- term frequency
- unlabeled data
- k nearest neighbor
- feature set
- dimensionality reduction
- support vector machine
- classification accuracy
- machine learning
- data mining
- linear svm
- text mining
- objective function
- learning algorithm