Having a blast : meta-learning and heterogeneous ensembles for data streams

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

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

31 Citations (Scopus)
1 Downloads (Pure)


Ensembles of classifiers are among the best performing classifiers available in many data mining applications. However, most ensembles developed specifically for the dynamic data stream setting rely on only one type of base-level classifier, most often Hoeffding Trees. In this paper, we study the use of heterogeneous ensembles, comprised of fundamentally different model types. Heterogeneous ensembles have proven successful in the classical batch data setting, however they do not easily transfer to the data stream setting. We therefore introduce the Online Performance Estimation framework, which can be used in data stream ensembles to weight the votes of (heterogeneous) ensemble members differently across the stream. Experiments over a wide range of data streams show performance that is competitive with state of the art ensemble techniques, including Online Bagging and Leveraging Bagging. All experimental results from this work are easily reproducible and publicly available on OpenML for further analysis.
Original languageEnglish
Title of host publication15th IEEE International Conference on Data Mining (ICDM 2015), 14-17 November 2015, Atlantic City, New Jersey
Place of PublicationPiscataway
PublisherInstitute of Electrical and Electronics Engineers
Number of pages6
ISBN (Electronic)978-1-4673-9504-5
ISBN (Print)978-1-4673-9503-8
Publication statusPublished - 7 Jan 2016
Event15th IEEE International Conference on Data Mining (ICDM 2015) - Atlantic City, United States
Duration: 14 Nov 201517 Nov 2015
Conference number: 15


Conference15th IEEE International Conference on Data Mining (ICDM 2015)
Abbreviated titleICDM 2015
Country/TerritoryUnited States
CityAtlantic City
Internet address


  • Data Streams
  • Ensembles
  • Meta-Learning


Dive into the research topics of 'Having a blast : meta-learning and heterogeneous ensembles for data streams'. Together they form a unique fingerprint.

Cite this