Nikunj C. Oza's Publications

Sorted by DateClassified by Publication TypeClassified by Research Category

Experimental Comparisons of Online and Batch Versions of Bagging and Boosting

Experimental Comparisons of Online and Batch Versions of Bagging and Boosting. Nikunj C. Oza and Stuart Russell. In The Seventh ACM SIGKDD International Conference on Knowledge Discovery and Data Mining, pp. 359–364, Morgan Kaufmann, San Francisco, California. USA, August 2001.

Download

[PDF]162.4kB  

Abstract

Bagging and boosting are well-known ensemble learning methods. They combine multiple learned base models with the aim of improving generalization performance. To date, they have been used primarily in batch mode, i.e., they require multiple passes through the training data. In previous work, we presented online bagging and boosting algorithms that only require one pass through the training data and presented experimental results on some relatively small datasets. Through additional experiments on a variety of larger synthetic and real datasets, this paper demonstrates that our online versions perform comparably to their batch counterparts in terms of classification accuracy. We also demonstrate the substantial reduction in running time we obtain with our online algorithms because they require fewer passes through the training data.

BibTeX Entry

@inproceedings{ozru01b,
	author={Nikunj C. Oza and Stuart Russell},
        title={Experimental Comparisons of Online and Batch Versions of Bagging and Boosting},
        booktitle={The Seventh ACM SIGKDD International Conference on Knowledge Discovery and Data Mining}
        publisher={Morgan Kaufmann},
        address={San Francisco, California. USA},
        editor={Foster Provost and Ramakrishnan Srikant},
        pages={359-364},
        month={August},
abstract={Bagging and boosting are well-known ensemble learning methods. They combine multiple learned base models with the aim of improving generalization performance. To date, they have been used primarily in batch mode, i.e., they require multiple passes through the training data. In previous work, we presented online bagging and boosting algorithms that only require one pass through the training data and presented experimental results on some relatively small datasets. Through additional experiments on a variety of larger synthetic and real datasets, this paper demonstrates that our online versions perform comparably to their batch counterparts in terms of classification accuracy. We also demonstrate the substantial reduction in running time we obtain with our online algorithms because they require fewer passes through the training data.},
bib2html_pubtype = {Refereed Conference},
bib2html_rescat = {Ensemble Learning},
        year = {2001}
}

Generated by bib2html.pl (written by Patrick Riley ) on Sun Jan 13, 2008 22:02:08