Anomaly-aware multiple instance learning for rare anemia disorder classification.
Salome KazeminiaArio SadafiAsya MakhroAnna BogdanovaShadi AlbarqouniCarsten MarrPublished in: CoRR (2022)
Keyphrases
- multiple instance learning
- image categorization
- supervised learning
- multiple instance
- class labels
- instance selection
- image set classification
- text classification
- classification accuracy
- feature vectors
- multi class
- semi supervised
- decision trees
- image annotation
- support vector
- object based image retrieval
- unsupervised learning
- image classification
- learning algorithm
- support vector machine
- feature extraction
- model selection
- training samples
- cross validation
- real valued
- image analysis
- density estimation
- training set
- training data
- high level
- image segmentation
- feature selection