Abstract
The effectiveness of the cost-sensitive listwise approach has been verified in learning to rank. However, the order preservation and generalization of cost-sensitive listwise approach are not studied. The two properties are very important since they can guide to develop a better ranking method. In this paper, we establish a framework for order preserved cost-sensitive listwise ranking approach. The framework yields the conditions of order preservation for the cost-sensitive listwise method. In addition, the generalization of the order preserved cost-sensitive listwise approach is proven. According to the theorem of generalization, a novel loss function for order preserved cost-sensitive listwise approach has been proposed. The loss function not only is order preserved but also penalizes the model complexity by an auxiliary variable. As an example, we propose the order preserved cost-sensitive ListMLE algorithm. Experimental results show the proposed method outperforms the baselines.
Access this chapter
Tax calculation will be finalised at checkout
Purchases are for personal use only
Preview
Unable to display preview. Download preview PDF.
Similar content being viewed by others
References
Cao, Z., Qin, T., Liu, T.Y., Tsai, M.F., Li, H.: Learning to rank: from pairwise approach to listwise approach. In: Proceedings of the 24th international Conference on Machine Learning, pp. 129–136. ACM, New York (2007)
Qin, T., Zhang, X.D., Tsai, M.F., Wang, D.S., Liu, T.Y., Li, H.: Query-level loss functions for information retrieval. Inf. Process. Manage. 44(2), 838–855 (2008)
Xia, F., Liu, T.Y., Wang, J., Zhang, W., Li, H.: Listwise approach to learning to rank: theory and algorithm. In: Proceedings of the 25th International Conference on Machine Learning, pp. 1192–1199. ACM, New York (2008)
Xia, F., Liu, T.Y., Li, H.: Top-k consistency of learning to rank methods. In: Advances in Neural Information Processing Systems, pp. 2098–2106 (2009)
Lu, M., Xie, M., Wang, Y., Liu, J., Huang, Y.: cost-sensitive listwise ranking approach. In: Proceedings of the 14th International Conference on Pacific-Asia Conference on Knowledge Discovery and Data Mining, pp. 358–366 (2010)
Marden, J.I.: Analyzing and Modeling Rank Data. Chapman and Hall, Boca Raton (1995)
Järvelin, K., Kekäläinen, J.: Cumulated gain-based evaluation of ir techniques. ACM Trans. Inf. Syst. 20(4), 422–446 (2002)
Chapelle, O., Wu, M.: Gradient descent optimization of smoothed information retrieval metrics. Information Retrieval
Boyd, S., Vandenberghe, L.: Convex Optimization. Cambridge University Press, Cambridge (March 2004)
Joachims, T.: Optimizing search engines using clickthrough data. In: Proceedings of the eighth ACM SIGKDD International Conference on Knowledge Discovery and Data Mining, pp. 133–142. ACM, New York (2002)
Xu, J., Li, H.: Adarank: a boosting algorithm for information retrieval. In: Proceedings of the 30th Annual International ACM SIGIR Conference on Research and Development in Information Retrieval, pp. 391–398. ACM, New York (2007)
Author information
Authors and Affiliations
Editor information
Editors and Affiliations
Rights and permissions
Copyright information
© 2010 Springer-Verlag Berlin Heidelberg
About this paper
Cite this paper
Lu, M., Xie, M., Wang, Y., Liu, J., Huang, Y. (2010). Order Preserved Cost-Sensitive Listwise Approach in Learning to Rank. In: Cheng, PJ., Kan, MY., Lam, W., Nakov, P. (eds) Information Retrieval Technology. AIRS 2010. Lecture Notes in Computer Science, vol 6458. Springer, Berlin, Heidelberg. https://doi.org/10.1007/978-3-642-17187-1_19
Download citation
DOI: https://doi.org/10.1007/978-3-642-17187-1_19
Publisher Name: Springer, Berlin, Heidelberg
Print ISBN: 978-3-642-17186-4
Online ISBN: 978-3-642-17187-1
eBook Packages: Computer ScienceComputer Science (R0)