Using graph-based metrics with empirical risk minimization to speed up active learning on networked data.
Sofus A. MacskassyPublished in: KDD (2009)
Keyphrases
- active learning
- networked data
- empirical risk minimization
- statistical learning theory
- semi supervised
- uniform convergence
- vc dimension
- sample complexity
- generalization error
- machine learning
- learning algorithm
- semi supervised learning
- random sampling
- relevance feedback
- supervised learning
- training set
- training examples
- labeled data
- phase transition
- unlabeled data
- support vector machine
- pairwise