Algorithm selection on data streams

J.R. Rijn, van, G. Holmes, B. Pfahringer, J. Vanschoren

Research output: Chapter in Book/Report/Conference proceedingConference contributionAcademicpeer-review

35 Citations (Scopus)


We explore the possibilities of meta-learning on data streams, in particular algorithm selection. In a first experiment we calculate the characteristics of a small sample of a data stream, and try to predict which classifier performs best on the entire stream. This yields promising results and interesting patterns. In a second experiment, we build a meta-classifier that predicts, based on measurable data characteristics in a window of the data stream, the best classifier for the next window. The results show that this meta-algorithm is very competitive with state of the art ensembles, such as OzaBag, OzaBoost and Leveraged Bagging. The results of all experiments are made publicly available in an online experiment database, for the purpose of verifiability, reproducibility and generalizability.
Original languageEnglish
Title of host publicationDiscovery Science (17th International Conference, DS 2014, Bled, Slovenia, October 8-10, 2014. Proceedings)
EditorsS. Dzeroski, P. Panov, D. Kocev, L. Todorovski
Place of PublicationHeidelberg
ISBN (Print)978-3-319-11811-6
Publication statusPublished - 2014
Eventconference; 17th International Conference on Discovery Science; 2014-10-08; 2014-10-10 -
Duration: 8 Oct 201410 Oct 2014

Publication series

NameLecture Notes in Computer Science
ISSN (Print)0302-9743


Conferenceconference; 17th International Conference on Discovery Science; 2014-10-08; 2014-10-10
Other17th International Conference on Discovery Science


Dive into the research topics of 'Algorithm selection on data streams'. Together they form a unique fingerprint.

Cite this