Handling outliers and concept drift in online mass flow prediction in CFB boilers.
Jorn BakkerMykola PechenizkiyIndre ZliobaiteAndriy IvannikovTommi KärkkäinenPublished in: KDD Workshop on Knowledge Discovery from Sensor Data (2009)
Keyphrases
- concept drift
- novelty detection
- data streams
- ensemble classifier
- drifting concepts
- streaming data
- non stationary
- prediction accuracy
- drift detection
- concept drifting
- stream clustering
- incremental clustering
- change detection
- data distribution
- data stream mining
- data stream classification
- classification algorithm
- evolving data streams
- prior knowledge
- outlier detection
- batch learning
- missing data
- data points
- incremental learning
- training data
- concept descriptions
- sliding window