Box drawings for learning with imbalanced data
Name
Rudin_Box drawings.pdf
Size
308.96 KB
Format
Adobe PDF
Checksum (MD5)
c3592955ced550114c084d6b4757bba7
Author(s)
Goh, Siong Thye
Rudin, Cynthia
Date Issued
August 2014
Journal
Proceedings of the 20th ACM SIGKDD international conference on Knowledge discovery and data mining (KDD '14)
Publisher
Association for Computing Machinery (ACM)
Citation
Siong Thye Goh and Cynthia Rudin. 2014. Box drawings for learning with imbalanced data. In Proceedings of the 20th ACM SIGKDD international conference on Knowledge discovery and data mining (KDD '14). ACM, New York, NY, USA, 333-342.
Version
Author's final manuscript
Abstract
The vast majority of real world classification problems are imbalanced, meaning there are far fewer data from the class of interest (the positive class) than from other classes. We propose two machine learning algorithms to handle highly imbalanced classification problems. The classifiers are disjunctions of conjunctions, and are created as unions of parallel axis rectangles around the positive examples, and thus have the benefit of being interpretable. The first algorithm uses mixed integer programming to optimize a weighted balance between positive and negative class accuracies. Regularization is introduced to improve generalization performance. The second method uses an approximation in order to assist with scalability. Specifically, it follows a \textit{characterize then discriminate} approach, where the positive class is characterized first by boxes, and then each box boundary becomes a separate discriminative classifier. This method has the computational advantages that it can be easily parallelized, and considers only the relevant regions of feature space.
MIT Department
Massachusetts Institute of Technology. Computer Science and Artificial Intelligence Laboratory
Massachusetts Institute of Technology. Operations Research Center
Sloan School of Management
Terms of Use
Creative Commons Attribution-Noncommercial-Share Alike
Persistent DSpace Link
DOI of Published Version
https://doi.org/10.1145/2623330.2623648
