Using diversity to handle concept drift in on-line learning

A recent study of diversity using on-line ensembles of learning machines on the presence of concept drift shows that different diversity levels are required before and after a drift. Besides, studies from the dynamic optimisation problems area suggest that, if the best solution for a particular time step is adopted, it may lead to a future scenario in which low accuracy is obtained. Based on that, we propose in this paper a new on-line ensemble learning approach to handle concept drift, which uses ensembles containing different diversity levels. Even though a high diversity ensemble may have low accuracy while the concept is stable, it may present better accuracy after a drift. The proposed approach successfully chooses the ensemble to be used when a concept drift occurs and shows to obtain better accuracy than a system which adopts the strategy of learning a new classifier from scratch when a drift is detected (strategy adopted by many of the current approaches that explicitly use a drift detection method).

[1]  Leo Breiman,et al.  Bagging Predictors , 1996, Machine Learning.

[2]  Stuart J. Russell,et al.  Experimental comparisons of online and batch versions of bagging and boosting , 2001, KDD '01.

[3]  Marcus A. Maloof,et al.  Using additive expert ensembles to cope with concept drift , 2005, ICML.

[4]  Stuart J. Russell,et al.  Online bagging and boosting , 2005, 2005 IEEE International Conference on Systems, Man and Cybernetics.

[5]  Paul E. Utgoff,et al.  Decision Tree Induction Based on Efficient Tree Restructuring , 1997, Machine Learning.

[6]  João Gama,et al.  Learning with Drift Detection , 2004, SBIA.

[7]  Xin Yao,et al.  The Impact of Diversity on Online Ensemble Learning in the Presence of Concept Drift , 2010, IEEE Transactions on Knowledge and Data Engineering.

[8]  M. Harries SPLICE-2 Comparative Evaluation: Electricity Pricing , 1999 .

[9]  Kyosuke Nishida,et al.  Adaptive Classifiers-Ensemble System for Tracking Concept Drift , 2007, 2007 International Conference on Machine Learning and Cybernetics.

[10]  Ludmila I. Kuncheva,et al.  A framework for generating data to simulate changing environments , 2007, Artificial Intelligence and Applications.

[11]  Vasant Honavar,et al.  Learn++: an incremental learning algorithm for supervised neural networks , 2001, IEEE Trans. Syst. Man Cybern. Part C.

[12]  Marcus A. Maloof,et al.  Dynamic weighted majority: a new ensemble method for tracking concept drift , 2003, Third IEEE International Conference on Data Mining.

[13]  Manfred K. Warmuth,et al.  The Weighted Majority Algorithm , 1994, Inf. Comput..

[14]  Xin Yao,et al.  Negative correlation in incremental learning , 2007, Natural Computing.

[15]  Koichiro Yamauchi,et al.  Detecting Concept Drift Using Statistical Testing , 2007, Discovery Science.

[16]  Robert Givan,et al.  Online Ensemble Learning: An Empirical Study , 2000, Machine Learning.

[17]  A. Bifet,et al.  Early Drift Detection Method , 2005 .

[18]  Xin Yao,et al.  On-line bagging Negative Correlation Learning , 2008, 2008 IEEE International Joint Conference on Neural Networks (IEEE World Congress on Computational Intelligence).

[19]  Ken Lang,et al.  NewsWeeder: Learning to Filter Netnews , 1995, ICML.

[20]  A. Dawid,et al.  Prequential probability: principles and properties , 1999 .

[21]  Nikola Kasabov,et al.  Evolving connectionist systems , 2002 .