Breast cancer identification: KDD CUP winner's report

ABSTRACT

We describe the ideas and methodologies that we developed in addressing the KDD Cup 2008 on early breast cancer detection, and discuss how they contributed to our success. The most important components of our solution were 1) the identification of predictive information in the patient identifier, 2) a linear SVM on the 117 provided features, and 3) a heuristic post-processing approach to optimize the evaluation criteria.

REFERENCES

• 1: Aron Inger , Nurit Vatnik , Saharon Rosset , Einat Neumann, KDD-Cup 2000: question 1 winner's report, ACM SIGKDD Explorations Newsletter, v.2 n.2, p.94, Dec. 2000 [doi>10.1145/380995.381038]
• 2: Thorsten Joachims, A support vector method for multivariate performance measures, Proceedings of the 22nd international conference on Machine learning, p.377-384, August 07-11, 2005, Bonn, Germany [doi>10.1145/1102351.1102399]
• 3: J. Platt. Probabilistic outputs for support vector machines and comparisons to regularized likelihood methods. In Advances in Large Margin Classifiers. MIT Press, 1998.
• 4: Saharon Rosset , Claudia Perlich , Yan Liu, Making the most of your data: KDD Cup 2007 "How Many Ratings" winner's report, ACM SIGKDD Explorations Newsletter, v.9 n.2, December 2007 [doi>10.1145/1345448.1345463]
• 5: G. Valentini and T.G. Dietterich. Low bias bagged support vector machines. In Proceedings of 20th International Conference on Machine Learning (ICML-2003), pages 752--759, Washington, DC, 2003.
• 6: R. Yan, J. Zhang, J. Yang, and A. Hauptmann. A discriminative learning framework with pairwise constraints for video object classification. Proceedings of IEEE Conference on Computer Vision and Pattern Recognition (CVPR'04), 2004.

INDEX TERMS

Primary Classification:
  A. General Literature
  A.0 GENERAL

General Terms:
Design, Documentation