Minimum rank error training for language modeling

Meng Sung Wu*, Jen-Tzung Chien

*Corresponding author for this work

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

Abstract

Discriminative training techniques have been successfully developed for many pattern recognition applications. In speech recognition, discriminative training aims to minimize the metric of word error rate. However, in an information retrieval system, the best performance should be achieved by maximizing the average precision. In this paper, we construct the discriminative n-gram language model for information retrieval following the metric of minimum rank error (MRE) rather than the conventional metric of minimum classification error. In the optimization procedure, we maximize the average precision and estimate the language model towards attaining the smallest ranking loss. In the experiments on ad-hoc retrieval using TREC collections, the proposed MRE language model performs better than the maximum likelihood and the minimum classification error language models.

Original languageEnglish
Title of host publicationInternational Speech Communication Association - 8th Annual Conference of the International Speech Communication Association, Interspeech 2007
Pages2700-2703
Number of pages4
StatePublished - 1 Dec 2007
Event8th Annual Conference of the International Speech Communication Association, Interspeech 2007 - Antwerp, Belgium
Duration: 27 Aug 200731 Aug 2007

Publication series

NameInternational Speech Communication Association - 8th Annual Conference of the International Speech Communication Association, Interspeech 2007
Volume4

Conference

Conference8th Annual Conference of the International Speech Communication Association, Interspeech 2007
CountryBelgium
CityAntwerp
Period27/08/0731/08/07

Keywords

  • Discriminative training
  • Information retrieval
  • Language model
  • Maximum average precision
  • Minimum rank error

Fingerprint Dive into the research topics of 'Minimum rank error training for language modeling'. Together they form a unique fingerprint.

  • Cite this

    Wu, M. S., & Chien, J-T. (2007). Minimum rank error training for language modeling. In International Speech Communication Association - 8th Annual Conference of the International Speech Communication Association, Interspeech 2007 (pp. 2700-2703). (International Speech Communication Association - 8th Annual Conference of the International Speech Communication Association, Interspeech 2007; Vol. 4).