On Theoretically Optimal Ranking Functions in Bipartite Ranking

@inproceedings{Uematsu2011OnTO,
  title={On Theoretically Optimal Ranking Functions in Bipartite Ranking},
  author={Kazuki Uematsu and Yoonkyung Lee},
  year={2011}
}
This paper investigates the theoretical relation between loss criteria and the optimal ranking functions driven by the criteria in bipartite ranking. In particular, the relation between AUC maximization and minimization of ranking risk under a convex loss is examined. We characterize general conditions for ranking-calibrated loss functions in a pairwise approach, and show that the best ranking functions under convex ranking-calibrated loss criteria produce the same ordering as the likelihood… CONTINUE READING

Citations

Publications citing this paper.
Showing 1-7 of 7 extracted citations

References

Publications referenced by this paper.
Showing 1-10 of 30 references

Generalization Bounds for the Area Under the ROC Curve

Journal of Machine Learning Research • 2005
View 5 Excerpts
Highly Influenced

Adapting ranking SVM

Y. Series. Cao, J. Xu, +5 authors H.-W
Machine learning, • 2006
View 4 Excerpts
Highly Influenced

All together now : A perspective on the Netflix prize

R. M. Bell, Y. Koren, C. Volinsky
Journal of the American Statistical Association • 2010
View 1 Excerpt

Composite Binary Losses

Journal of Machine Learning Research • 2010

Similar Papers

Loading similar papers…