Parametric methods for comparing the performance of two classification algorithms evaluated by k-fold cross validation on multiple data sets.
Tzu-Tsung WongPublished in: Pattern Recognit. (2017)
Keyphrases
- benchmark datasets
- machine learning methods
- benchmark data sets
- data sets
- significant improvement
- machine learning algorithms
- computational cost
- methods outperform
- preprocessing
- feature selection algorithms
- high dimensional data
- terms of classification accuracy
- methods require
- classification systems
- classification method
- classification algorithm
- data mining techniques
- machine learning
- optimization methods
- computationally expensive
- real world data sets
- synthetic and real datasets
- uci machine learning repository
- search methods
- supervised learning tasks
- feature extraction and selection
- publicly available data sets
- small data sets
- computational complexity
- classification accuracy
- text classification
- cross validation
- density estimation
- supervised learning
- induction algorithms
- data reduction
- pattern recognition
- training data
- decision trees
- feature selection
- supervised classifiers
- feature subset