Classification accuracy improvement of neural network classifiers by using unlabeled data.
M. T. FardaneshOkan K. ErsoyPublished in: IEEE Trans. Geosci. Remote. Sens. (1998)
Keyphrases
- unlabeled data
- classification accuracy
- labeled data
- semi supervised learning
- training data
- training set
- semi supervised
- improve the classification accuracy
- active learning
- co training
- supervised learning
- data points
- labeled training data
- semi supervised classification
- text classification
- labeled examples
- labeled and unlabeled data
- supervised learning algorithms
- feature selection
- learning algorithm
- text categorization
- training examples
- small set of labeled
- prior knowledge
- domain adaptation
- class labels
- positive examples
- data sets
- feature space
- support vector
- multi view learning
- semisupervised learning
- naive bayes
- unsupervised learning
- multi view
- feature set
- pairwise
- label propagation
- labeled instances
- labeled data for training
- class distribution
- transfer learning
- kernel function
- training samples
- information extraction
- decision trees
- information retrieval
- machine learning