Boosting rank with predictable training error

Yongqing Wang, Wenji Mao, Daniel Zeng, Ning Bao

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Abstract

Listwise approach is an important method to solve practical Web search problem in learning to rank. In this paper, we first analyze the practical Web search problem and construct the model to solve it. Then we propose an algorithm called DiffRank which can apply boosting technology to learning to rank in listwise. Through theoretical analysis, we prove that the upper bound of training error can be reduced in our proposed algorithm. The experimental results further verify our theoretical analysis and demonstrate that our approach can better perform in practical Web search than other state-of-the-art listwise algorithms.

Original languageEnglish (US)
Title of host publicationProceedings of 2011 IEEE International Conference on Intelligence and Security Informatics, ISI 2011
Pages407-409
Number of pages3
DOIs
StatePublished - Sep 22 2011
Event2011 IEEE International Conference on Intelligence and Security Informatics, ISI 2011 - Beijing, China
Duration: Jul 10 2011Jul 12 2011

Publication series

NameProceedings of 2011 IEEE International Conference on Intelligence and Security Informatics, ISI 2011

Other

Other2011 IEEE International Conference on Intelligence and Security Informatics, ISI 2011
CountryChina
CityBeijing
Period7/10/117/12/11

Keywords

  • DiffRank
  • boosting
  • learning to rank

ASJC Scopus subject areas

  • Artificial Intelligence
  • Information Systems

Cite this

Wang, Y., Mao, W., Zeng, D., & Bao, N. (2011). Boosting rank with predictable training error. In Proceedings of 2011 IEEE International Conference on Intelligence and Security Informatics, ISI 2011 (pp. 407-409). [5984123] (Proceedings of 2011 IEEE International Conference on Intelligence and Security Informatics, ISI 2011). https://doi.org/10.1109/ISI.2011.5984123