Classification from Pairwise Similarities/Dissimilarities and Unlabeled Data via Empirical Risk Minimization.
Takuya ShimadaHan BaoIssei SatoMasashi SugiyamaPublished in: CoRR (2019)
Keyphrases
- unlabeled data
- co training
- semi supervised learning
- labeled data
- supervised learning
- labeled and unlabeled data
- text classification
- semi supervised
- class labels
- training set
- active learning
- learning algorithm
- training data
- decision boundary
- classification accuracy
- machine learning
- pairwise
- text categorization
- unsupervised learning
- pattern recognition
- training examples
- feature vectors
- transfer learning
- feature selection
- support vector machine
- decision trees
- training samples
- data points
- support vector
- supervised classification
- feature extraction
- domain adaptation
- feature space
- naive bayes
- nearest neighbor
- prior knowledge
- svm classifier
- euclidean space
- model selection