Paired Learners for Concept Drift
Bach, S.H.
Maloof, M.A.
Dept. of Comput. Sci., Georgetown Univ., Washington, DC;
This paper appears in: Data Mining, 2008. ICDM '08. Eighth IEEE International Conference on
Publication Date: 15-19 Dec. 2008
On page(s): 23-32
Location: Pisa,
ISSN: 1550-4786
ISBN: 978-0-7695-3502-9
INSPEC Accession Number: 10472139
Digital Object Identifier: 10.1109/ICDM.2008.119
Current Version Published: 2009-02-10
Abstract
To cope with concept drift, we paired a stable online learner with a reactive one. A stable learner predicts based on all of its experience, whereas are active learner predicts based on its experience over a short, recent window of time. The method of paired learning uses differences in accuracy between the two learners over this window to determine when to replace the current stable learner, since the stable learner performs worse than does there active learner when the target concept changes. While the method uses the reactive learner as an indicator of drift, it uses the stable learner to predict, since the stable learner performs better than does the reactive learner when acquiring target concept. Experimental results support these assertions. We evaluated the method by making direct comparisons to dynamic weighted majority, accuracy weighted ensemble, and streaming ensemble algorithm (SEA) using two synthetic problems, the Stagger concepts and the SEA concepts, and three real-world data sets: meeting scheduling, electricity prediction, and malware detection. Results suggest that, on these problems, paired learners outperformed or performed comparably to methods more costly in time and space.
Index
Terms
Available to subscribers and IEEE members.
References
Available to subscribers and IEEE members.
Citing Documents
Available to subscribers and IEEE members.
Cart
