A method on selecting reliable samples based on fuzziness in positive and unlabeled learning.
TingTing LiWeiYa FanYunSong LuoPublished in: CoRR (2019)
Keyphrases
- prior knowledge
- unsupervised learning
- training samples
- similarity measure
- high accuracy
- preprocessing
- detection method
- significant improvement
- data sets
- supervised learning
- segmentation method
- computational complexity
- pairwise
- learning algorithm
- fuzzy logic
- learning systems
- clustering method
- learning mechanism
- computational cost
- support vector machine svm
- classification accuracy
- active learning
- learning environment
- reinforcement learning
- neural network