An approach to handle concept drift in financial time series based on Extreme Learning Machines and explicit Drift Detection.
Rodolfo C. CavalcanteAdriano L. I. OliveiraPublished in: IJCNN (2015)
Keyphrases
- drift detection
- concept drift
- financial time series
- non stationary
- extreme learning machines
- stock market
- data streams
- drifting concepts
- streaming data
- change detection
- exchange rate
- financial data
- data stream mining
- classification algorithm
- stock price
- data distribution
- multivariate time series
- long run
- text categorization
- image processing