AdaRank on LETOR 

 

Introduction

Papers&Docs

Notes

 

Introduction to AdaRank

The basic idea of AdaRank is repeatedly constructing 'weak rankers' on the basis of reweighted training queries and finally linearly combining the weak rankers for making ranking predictions. AdaRank-MAP utilizes MAP as performance measure function to measure the goodness of a weak ranker, while AdaRank-NDCG utilizes NDCG.

The details of AdaRank can be found from SIGIR2007 paper.

 

Papers & Docs

Jun Xu and Hang Li. AdaRank: A Boosting Algorithm for Information Retrieval, SIGIR'07: Proceedings of the 30th annual international ACM SIGIR conference on Research and development in information retrieval.

BibTex
@inproceedings{1277809, 
   author = {Jun Xu and Hang Li},
   title = {AdaRank: a boosting algorithm for information retrieval}, 
   booktitle = {SIGIR '07: Proceedings of the 30th annual international ACM SIGIR conference on Research and development in information retrieval},
   year = {2007}, 
   isbn = {978-1-59593-597-7}, 
   pages = {391--398}, 
   location = {Amsterdam, The Netherlands}, 
   doi = {http://doi.acm.org/10.1145/1277741.1277809}, 
   publisher = {ACM}, 
   address = {New York, NY, USA},
}  
 

Notes

This document was written by Jun Xu and the experiments were conducted by Chaoliang Zhong.  If any problem, please contact letor@microsoft.com.