The Majority Can Help the Minority: Context-rich Minority Oversampling for Long-tailed Classification.
Seulki ParkYoungkyu HongByeongho HeoSangdoo YunJin Young ChoiPublished in: CVPR (2022)
Keyphrases
- minority class
- majority class
- class imbalance
- class imbalanced
- decision trees
- class distribution
- classification scheme
- support vector machine
- pattern recognition
- imbalanced datasets
- machine learning
- classification process
- automatic classification
- classification accuracy
- classification error
- classification method
- high dimensionality
- decision boundary
- feature vectors
- cost sensitive
- machine learning methods
- benchmark datasets
- support vector machine svm
- contextual information
- text classification
- supervised learning
- real world
- preprocessing
- feature selection
- attribute selection
- highly imbalanced
- feature space
- active learning
- original data
- machine learning algorithms
- context aware
- image classification