Boosted Near-miss Under-sampling on SVM ensembles for concept detection in large-scale imbalanced datasets.
Lei BaoCao JuanJintao LiYongdong ZhangPublished in: Neurocomputing (2016)
Keyphrases
- imbalanced datasets
- concept detection
- imbalanced data
- ensemble methods
- random forests
- decision trees
- video retrieval
- class distribution
- class imbalance
- cost sensitive learning
- prediction accuracy
- sampling methods
- concept learning
- training dataset
- support vector machine
- generalization ability
- support vector
- semantic concepts
- ensemble learning
- random forest
- base classifiers
- ensemble classifier
- benchmark datasets
- support vector machine svm
- base learners
- feature selection
- machine learning methods
- cost sensitive
- binary classification
- feature selection algorithms
- minority class
- feature vectors
- video data
- training set
- nearest neighbor
- multi class