• Search Research Projects
  • Search Researchers
  • How to Use
  1. Back to previous page

A leap from short range language models to middle range modeling using dependency ngrams

Research Project

Project/Area Number 24650063
Research Category

Grant-in-Aid for Challenging Exploratory Research

Allocation TypeSingle-year Grants
Research Field Intelligent informatics
Research InstitutionUniversity of Tsukuba

Principal Investigator

YAMAMOTO MIKIO  筑波大学, システム情報系, 教授 (40210562)

Project Period (FY) 2012-04-01 – 2014-03-31
Project Status Completed (Fiscal Year 2013)
Budget Amount *help
¥3,770,000 (Direct Cost: ¥2,900,000、Indirect Cost: ¥870,000)
Fiscal Year 2013: ¥1,820,000 (Direct Cost: ¥1,400,000、Indirect Cost: ¥420,000)
Fiscal Year 2012: ¥1,950,000 (Direct Cost: ¥1,500,000、Indirect Cost: ¥450,000)
Keywords確率的言語モデル / 依存構造 / 機械翻訳 / 言語モデル / 構文解析 / EMアルゴリズム
Research Abstract

Statistical language models are a fundamental component of speech recognition systems, machine translation systems, and so forth. Presently, the ngram language model is the most widely used approach. This model focuses on sequences of neighboring lexical words, and uses the probabilities of these sequences as model parameters. Due to the full lexicalization of the ngram language model, local features of word sequences can be well modeled. However, an ngram language model cannot capture relatively medium or long-range features, because it regards a sentence as a flat string and ignores its structure. In this research, we proposed a generative dependency ngram language model that integrates a generative dependency structure of a sentence into the original ngram language model. Using an expectation-maximization (EM) algorithm, the probability of arbitrary order dependency ngrams can be estimated by considering all possible dependency structures of a sentence.

Report

(3 results)
  • 2013 Annual Research Report   Final Research Report ( PDF )
  • 2012 Research-status Report
  • Research Products

    (5 results)

All 2014 2013

All Journal Article (1 results) (of which Peer Reviewed: 1 results) Presentation (4 results)

  • [Journal Article] A generative dependency N-gram language model : unsupervised parameter estimation and application2014

    • Author(s)
      Chenchen Ding and Mikio Yamamoto
    • Journal Title

      自然言語処理

      Volume: Vol.21, No.5(印刷予定)

    • NAID

      130004705299

    • Related Report
      2013 Final Research Report
    • Peer Reviewed
  • [Presentation] An unsupervised parameter estimation algorithm for a generative dependency N-gram language model2013

    • Author(s)
      Chenchen Ding and Mikio Yamamoto
    • Organizer
      In Proc. of IJCNLP 2013
    • URL

      http://lang.cs.tut.ac.jp/ijcnlp2013/

    • Related Report
      2013 Final Research Report
  • [Presentation] An Unsupervised Parameter Estimation Algorithm for a Generative Dependency N-gram Language Model2013

    • Author(s)
      Chenchen Ding and Mikio Yamamoto
    • Organizer
      The 6th International Joint Conference on Natural Language Processing
    • Place of Presentation
      Nagoya, Japan
    • Related Report
      2013 Annual Research Report
  • [Presentation] An efficient language model using double-array structures2013

    • Author(s)
      M.Yasuhara, T.Tanaka, J.Norimatsu and M.Yamamoto
    • Organizer
      The 2013 Conference on Empirical Methods in Natural Language Processing
    • Place of Presentation
      Seattle, USA
    • Related Report
      2013 Annual Research Report
  • [Presentation] Double-Arrayを利用した高速かつコンパクトなngram言語モデルの構築手法2013

    • Author(s)
      安原誠, 田中透, 乗松潤矢, 山本幹雄
    • Organizer
      言語処理学会第19回年次大会
    • Place of Presentation
      名古屋大学東山キャンパス(愛知県)
    • Related Report
      2012 Research-status Report

URL: 

Published: 2013-05-31   Modified: 2019-07-29  

Information User Guide FAQ News Terms of Use Attribution of KAKENHI

Powered by NII kakenhi