cpp-ToyBox-Ranking

Now, under developing. There are insufficient unit tests.

This library is a tiny package for learning-to-rank problems. This library currently supports:

• RankSVM [1]
• RankNet [2,7]
• ListNet [3]
• ListMLE [4]
• LambdaRank [5,7]
• LambdaMART [6,7]

These implemented models are currently linear model, except for LambdaMART. No non-linear kernel in SVM, no hidden layer in neural networks.

REFERENCES

• [1] T. Joachims. (2002). Optimizing Search Engines Using Clickthrough Data. KDD 2002.
• [2] C. Burges, T. Shaked, E. Renshaw, A. Lazier, M. Deeds, N. Hamilton, and G. Hullender. (2005). Learning to Rank using Gradient Descent. ICML 2005.
• [3] Z. Cao, T. Qin, T.-Y. Liu, M.-F. Tsai, and H. Li. (2007). Learning to Rank: From Pairwise Approach to Listwise Approach. ICML 2007.
• [4] F. Xia, T.-Y. Liu, J. Wang, W. Zhang, and H. Li. (2008). Listwise Approach to Learning to Rank - Theory and Algorithm. ICML 2008.
• [5] C. J. C. Burges, R. Ragno, and Q. V. Le. (2006). Learning to Rank with Nonsmooth Cost Functions. NIPS 2006.
• [6] Q. Wu, C. Burges, K. Svore, and J. Gao. (2008). Ranking, Boosting and Model Adaptation. Microsoft Technical Report MSR-TR-2008-109.
• [7] C. J.C. Burges. (2010). From RankNet to LambdaRank to LambdaMART: An Overview. Microsoft Research Technical Report MSR-TR-2010-82.

LINKS

• RankLib
• jforests

AUTHOR

TAGAMI Yukihiro [email protected]

LICENSE

This library is distributed under the term of the MIT license. http://opensource.org/licenses/mit-license.php