Exploiting Unlabeled Data for Improving Accuracy of Predictive Data Mining.
Kang PengSlobodan VuceticBo HanHongbo M. XieZoran ObradovicPublished in: ICDM (2003)
Keyphrases
- unlabeled data
- labeled data
- semi supervised learning
- semi supervised
- predictive data mining
- active learning
- co training
- supervised learning
- training data
- labeled instances
- text classification
- labeled and unlabeled data
- data points
- training set
- training examples
- class labels
- text categorization
- learning algorithm
- machine learning
- unsupervised learning
- intelligent search
- pairwise
- classification accuracy
- class distribution
- decision boundary
- pairwise constraints