A two-step anomaly detection based method for PU classification in imbalanced data sets.
Carlos Ortega VázquezSeppe vanden BrouckeJochen De WeerdtPublished in: Data Min. Knowl. Discov. (2023)
Keyphrases
- anomaly detection
- unsupervised learning
- support vector machine svm
- benchmark data sets
- classification accuracy
- support vector machine
- one class support vector machines
- training samples
- pattern classification
- imbalanced data sets
- text classification
- pattern recognition
- classification algorithm
- pairwise
- support vector
- similarity measure
- data sets