Recognizing input space and target concept drifts in data streams with scarcely labeled and unlabelled instances.
Edwin LughoferEva WeiglWolfgang HeidlChristian EitzingerThomas RadauerPublished in: Inf. Sci. (2016)
Keyphrases
- input space
- concept drift
- data streams
- data stream classification
- class labels
- streaming data
- training set
- classification algorithm
- data stream mining
- dimensionality reduction
- mining data streams
- feature space
- input data
- kernel function
- sliding window
- k nearest neighbor
- stream data
- training data
- change detection
- evolving data streams
- data points
- hyperplane
- drift detection
- non stationary
- low dimensional
- high dimensional
- supervised learning
- data distribution
- concept drifting data streams
- knn
- target domain
- stream mining
- high dimensional data
- itemsets
- data sets
- semi supervised learning
- labeled data
- continuous queries
- nearest neighbor
- pattern recognition
- unlabeled data
- stream processing
- support vector machine
- concept drifting
- feature extraction