Conference proceeding
Learning to rank by maximizing AUC with linear programming
2006 IEEE INTERNATIONAL JOINT CONFERENCE ON NEURAL NETWORK PROCEEDINGS, VOLS 1-10, pp.123-129
IEEE International Joint Conference on Neural Networks (IJCNN)
01/01/2006
DOI: 10.1109/IJCNN.2006.246669
Abstract
Area Under the ROC Curve (AUC) is often used to evaluate ranking performance in binary classification problems. Several researchers have approached AUC optimization by approximating the equivalent Wicoxon-Mann-Whitney (WMW) statistic. We present a linear programming approach similar to 1-norm Support Vector Machines (SVMs) for instance ranking by an approximation to the WMW statistic. Our formulation can be applied to nonlinear problems by using a kernel function. Our ranking algorithm outperforms SVMs in both AUC and classification performance when using RBF kernels, but curiously not with polynomial kernels. We experiment with variations of chunking to handle the quadratic growth of the number of constraints in our formulation.
Details
- Title: Subtitle
- Learning to rank by maximizing AUC with linear programming
- Creators
- Kaan Ataman - University of IowaW. Nick Street - University of IowaYi Zhang - University of Iowa
- Resource Type
- Conference proceeding
- Publication Details
- 2006 IEEE INTERNATIONAL JOINT CONFERENCE ON NEURAL NETWORK PROCEEDINGS, VOLS 1-10, pp.123-129
- Publisher
- IEEE
- Series
- IEEE International Joint Conference on Neural Networks (IJCNN)
- DOI
- 10.1109/IJCNN.2006.246669
- ISSN
- 2161-4393
- eISSN
- 2161-4407
- Number of pages
- 3
- Language
- English
- Date published
- 01/01/2006
- Academic Unit
- Nursing; Business Analytics; Computer Science; Bus Admin College
- Record Identifier
- 9984380522702771
Metrics
1 Record Views