Exploiting unlabeled data to improve peer-to-peer traffic classification using incremental tri-training method.
Bijan RaahemiWeicai ZhongJing LiuPublished in: Peer-to-Peer Netw. Appl. (2009)
Keyphrases
- unlabeled data
- improve the classification accuracy
- training set
- semisupervised learning
- training process
- supervised learning
- semi supervised learning
- semi supervised classification
- classification algorithm
- class labels
- unsupervised learning
- training samples
- text classification
- support vector machine
- classification accuracy
- pattern classification
- semi supervised
- active learning
- prior knowledge
- supervised learning algorithms
- decision boundary
- model selection
- training examples
- peer to peer
- machine learning
- similarity measure
- select relevant features
- co training
- labeled data
- feature vectors
- training data