COMB: A Hybrid Method for Cross-validated Feature Selection.
G. S. ThejasDaniel JimenezS. Sitharama IyengarJerry MillerN. R. SunithaPrajwal BadrinathPublished in: ACM Southeast Regional Conference (2020)
Keyphrases
- hybrid method
- cross validated
- feature selection
- fold cross validation
- cross validation
- support vector machine
- quantile regression
- log likelihood
- model selection
- support vector
- text categorization
- machine learning
- logistic regression
- hybrid algorithm
- hyperparameters
- mutual information
- text classification
- feature extraction
- classification accuracy
- multi class
- feature subset selection
- knn
- feature space
- similarity measure
- feature subset
- neural network