Issue No. 06 - November/December (2008 vol. 12)
DOI Bookmark: http://doi.ieeecomputersociety.org/10.1109/MIC.2008.119
Jing Gao , University of Illinois, Urbana-Champaign
Bolin Ding , University of Illinois, Urbana-Champaign
Wei Fan , IBM T.J. Watson Research Center
Jiawei Han , University of Illinois, Urbana-Champaign
Philip S. Yu , University of Illinois, Chicago
Classification is an important data analysis tool that uses a model built from historical data to predict class labels for new observations. More and more applications are featuring data streams, rather than finite stored data sets, which are a challenge for traditional classification algorithms. Concept drifts and skewed distributions, two common properties of data stream applications, make the task of learning in streams difficult. The authors aim to develop a new approach to classify skewed data streams that uses an ensemble of models to match the distribution over under-samples of negatives and repeated samples of positives.
data stream, classification algorithms, concept drifts, data mining, model averaging, skewed distributions
J. Gao, P. S. Yu, W. Fan, B. Ding and J. Han, "Classifying Data Streams with Skewed Class Distributions and Concept Drifts," in IEEE Internet Computing, vol. 12, no. , pp. 37-49, 2008.