MixMOOD: A systematic approach to class distribution mismatch in semi-supervised learning using deep dataset dissimilarity measures.
Saúl Calderón RamírezLuis OalaJordina Torrents-BarrenaShengxiang YangArmaghan MoemeniWojciech SamekMiguel A. Molina-CabelloPublished in: CoRR (2020)
Keyphrases
- semi supervised learning
- class distribution
- unlabeled data
- dissimilarity measure
- training data
- labeled data
- semi supervised
- class imbalance
- supervised learning
- training set
- training dataset
- machine learning
- active learning
- feature space
- text categorization
- training samples
- training examples
- distance measure
- test data
- cost sensitive
- class labels
- test set
- learning algorithm
- benchmark datasets
- unsupervised learning
- data points
- similarity measure
- distance function
- decision boundary
- clustering method
- text classification
- concept drift
- prior knowledge
- clustering algorithm
- naive bayes
- classification accuracy
- feature vectors