Evaluating boosting algorithms to classify rare classes: Comparison and improvements

Mahesh V. Joshi, Vipin Kumar, Ramesh C. Agarwal

Research output: Chapter in Book/Report/Conference proceedingConference contribution

161 Scopus citations

Abstract

Classification of rare events has many important data mining applications. Boosting is a promising meta-technique that improves the classification peformance of any weak classifier. So far, no systematic study has been conducted to evaluate how boosting performs for the task of mining rare classes. In this paper, we evaluate three existing categories of boosting algorithms from the single viewpoint of how they update the example weights in each iteration, and discuss their possible effect on recall and precision of the rare class. We propose enhanced algorithms in two of the categories, arid justib their choice of weight updating parameters theoretically. Using some specially designed synthetic datasets, we compare the capability of all the algorithms from the rare class perspective. The results support our qualitative analysis, and also indicate that our enhancements bring an extra capability for achieving better balance between recall and precision in mining rare classes.

Original languageEnglish (US)
Title of host publicationProceedings - 2001 IEEE International Conference on Data Mining, ICDM'01
Pages257-264
Number of pages8
StatePublished - Dec 1 2001
Event1st IEEE International Conference on Data Mining, ICDM'01 - San Jose, CA, United States
Duration: Nov 29 2001Dec 2 2001

Publication series

NameProceedings - IEEE International Conference on Data Mining, ICDM
ISSN (Print)1550-4786

Other

Other1st IEEE International Conference on Data Mining, ICDM'01
Country/TerritoryUnited States
CitySan Jose, CA
Period11/29/0112/2/01

Fingerprint

Dive into the research topics of 'Evaluating boosting algorithms to classify rare classes: Comparison and improvements'. Together they form a unique fingerprint.

Cite this