Facing the reality of data stream classification: coping with scarcity of labeled data.
Mohammad M. MasudClay WoolamJing GaoLatifur KhanJiawei HanKevin W. HamlenNikunj C. OzaPublished in: Knowl. Inf. Syst. (2011)
Keyphrases
- labeled data
- data stream classification
- concept drift
- unlabeled data
- semi supervised
- data streams
- semi supervised learning
- training data
- active learning
- text classification
- supervised learning
- data stream mining
- streaming data
- incremental learning
- data points
- data distribution
- semi supervised classification
- transfer learning
- non stationary
- prior knowledge
- classification algorithm
- class labels
- labeled training data
- learning algorithm
- supervised methods
- domain adaptation
- change detection
- training examples
- labeled examples
- sliding window
- co training
- feature selection
- labeled and unlabeled data
- missing values
- knn
- database
- nearest neighbor
- training set
- learning tasks
- machine learning
- data mining
- labeled instances
- data sets