Probability model selection and parameter evolutionary estimation for clustering imbalanced data without sampling.
Jiancong FanZhonghan NiuYongquan LiangZhongying ZhaoPublished in: Neurocomputing (2016)
Keyphrases
- model selection
- imbalanced data
- feature selection
- parameter estimation
- cross validation
- sampling methods
- unsupervised learning
- sample size
- hyperparameters
- linear regression
- clustering algorithm
- ensemble methods
- class distribution
- probability distribution
- decision trees
- high dimensionality
- machine learning
- support vector machine
- meta learning
- class imbalance
- classification accuracy
- generalization error
- random sampling
- random forest
- parameter space
- feature set