A general optimization framework for smoothing language models on graph structures

Qiaozhu Mei, Duo Zhang, Chengxiang Zhai

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

Abstract

Recent work on language models for information retrieval has shown that smoothing language models is crucial for achieving good retrieval performance. Many different effective smoothing methods have been proposed, which mostly implement various heuristics to exploit corpus structures. In this paper, we propose a general and unified optimization framework for smoothing language models on graph structures. This framework not only provides a unified formulation of the existing smoothing heuristics, but also serves as a road map for systematically exploring smoothing methods for language models. We follow this road map and derive several different instantiations of the framework. Some of the instantiations lead to novel smoothing methods. Empirical results show that all such instantiations are effective with some outperforming the state of the art smoothing methods.

Original languageEnglish (US)
Title of host publicationACM SIGIR 2008 - 31st Annual International ACM SIGIR Conference on Research and Development in Information Retrieval, Proceedings
Pages611-618
Number of pages8
DOIs
StatePublished - 2008
Event31st Annual International ACM SIGIR Conference on Research and Development in Information Retrieval, ACM SIGIR 2008 - Singapore, Singapore
Duration: Jul 20 2008Jul 24 2008

Publication series

NameACM SIGIR 2008 - 31st Annual International ACM SIGIR Conference on Research and Development in Information Retrieval, Proceedings

Other

Other31st Annual International ACM SIGIR Conference on Research and Development in Information Retrieval, ACM SIGIR 2008
Country/TerritorySingapore
CitySingapore
Period7/20/087/24/08

Keywords

  • Document and word graph
  • Graph structure
  • Language modeling
  • Smoothing

ASJC Scopus subject areas

  • Information Systems
  • Software

Fingerprint

Dive into the research topics of 'A general optimization framework for smoothing language models on graph structures'. Together they form a unique fingerprint.

Cite this