Hauptnavigation

Software developed at LS8

Several progams have been and are developed in parallel to our research activity. Most of them are actively maintained by their author.
TitelDescriptionAuthor(s)Related projectsID
ADT Hueppe, Christian
du6pki44
A Drift-based Dynamic Ensemble Members Selection using Clustering For Time Series ForecastingThis package enables a dynamic selection of heteregeneous ensemble base models through a performance drift detection mechanism and ensures ensemble diversity through a second stage selection using clustering that is computed after each drift detection. Predictions of final selected models are combined single prediction using sliding-window averages or stacking. Saadallah, Amal
SFB 876
g8c6g7ec5c