OLINDDA: a cluster-based approach for detecting novelty and concept drift in data streams.
Eduardo Jaques SpinosaAndré Carlos Ponce de Leon Ferreira de CarvalhoJoão GamaPublished in: SAC (2007)
Keyphrases
- concept drift
- data streams
- novelty detection
- streaming data
- stream clustering
- data stream classification
- mining data streams
- drift detection
- evolving data streams
- change detection
- sliding window
- drifting concepts
- batch learning
- stream data
- data stream mining
- data distribution
- data points
- high speed data streams
- clustering algorithm
- ensemble classifier
- classification algorithm
- non stationary
- continuous queries
- incremental clustering
- data clustering
- stream mining
- text mining
- neural network
- stream processing
- incoming data
- outlier detection
- k nearest neighbor
- small number