Semi-supervised active learning algorithm for SVMs based on QBC and tri-training.
Hailong XuLongyue LiPengsong GuoPublished in: J. Ambient Intell. Humaniz. Comput. (2021)
Keyphrases
- semi supervised
- supervised learning
- batch mode
- learning algorithm
- training algorithm
- training examples
- unlabeled data
- support vector
- active learning
- training set
- labeled data
- training samples
- learning machines
- generalization ability
- semi supervised learning
- training data
- partially labeled data
- training process
- support vectors
- fully labeled
- machine learning
- maximum margin
- svm training
- support vector machine
- query by committee
- feature selection
- early stopping
- co training
- pairwise
- back propagation
- training procedure
- learning problems
- training support vector machines
- structured learning
- restricted boltzmann machine
- training speed
- knn
- stochastic gradient descent
- structured prediction
- neural network
- classification algorithm
- loss function
- labeled examples
- feature vectors
- decision boundary
- multi class
- learning rate
- learning tasks
- svm classifier
- reduced set
- artificial neural networks
- text classification
- support vector machine svm
- learning models
- hyperplane
- cross validation