Mitigating Overfitting in Supervised Classification from Two Unlabeled Datasets: A Consistent Risk Correction Approach.
Nan LuTianyi ZhangGang NiuMasashi SugiyamaPublished in: CoRR (2019)
Keyphrases
- supervised classification
- supervised learning
- unsupervised learning
- risk management
- supervised classifiers
- active learning
- statistical learning theory
- unsupervised classification
- unlabeled data
- land cover classification
- unsupervised clustering
- cross validation
- remotely sensed images
- machine learning
- labeled data
- benchmark datasets
- data swapping
- risk assessment
- semi supervised learning
- decision trees
- decision making
- decision support system
- semi supervised
- model averaging
- data sets
- database
- data clustering
- land cover
- bayesian classifiers