A high-quality feature selection method based on frequent and correlated items for text classification.
Heba Mamdouh FarghalyTarek Abd-El-HafeezPublished in: Soft Comput. (2023)
Keyphrases
- text classification
- feature selection
- high quality
- significant improvement
- similarity measure
- feature set
- method for feature selection
- data mining
- classification method
- preprocessing
- detection method
- pairwise
- classification accuracy
- text mining
- probabilistic model
- information retrieval
- neural network
- support vector machine
- dimensionality reduction
- data sets
- mutual information
- feature space
- feature subset
- bayes classifier