On the importance of the validation technique for classification with imbalanced datasets: Addressing covariate shift when data is skewed.
Victoria LópezAlberto FernándezFrancisco HerreraPublished in: Inf. Sci. (2014)
Keyphrases
- highly skewed
- data distribution
- data sets
- imbalanced datasets
- training data
- original data
- small number
- data analysis
- training set
- input data
- training samples
- decision trees
- classification accuracy
- data sources
- reinforcement learning
- high dimensional data
- benchmark datasets
- sliding window
- feature extraction
- missing values
- class imbalance
- training dataset
- learning algorithm
- machine learning