An Evaluation of a Layered Neural Network which Have Function of Learning Vectorial Symbol Representations on PP-attachment Ambiguity Resolution

DOI IR (HANDLE) Web Site Open Access
  • Motoki Minoru
    Department of Intelligent Systems, Graduate School of Information Science and Electrical Engineering, Kyushu University : Doctoral Program
  • Tomiura Yoichi
    Department of Intelligent Systems, Faculty of Information Science and Electrical Engineering, Kyushu University
  • Hitaka Toru
    Professor Emeritus, Kyushu University
  • Shimazu Yoshio
    Department of Electrical Engineering, School of Engineering, Kyushu Sangyo University
  • Takahashi Naoto
    National Institute of Advanced Industrial Science and Technology

Bibliographic Information

Other Title
  • 記号表現ベクトル学習機能を有するニューラルネットの英語前置詞句係り先決定問題における実験的評価
  • キゴウ ヒョウゲン ベクトル ガクシュウ キノウ オ ユウスル ニューラルネット ノ エイゴ ゼンチシク カカリ サキ ケッテイ モンダイ ニ オケル ジッケンテキ ヒョウカ

Search this article

Description

This paper describes a PP-attachment ambiguity resolution with a layered neural network which have function of learning vectorial symbol representations. The proposed model does not update only link weight but also vectorial symbol representations. We show qualitative difference between a proposed model and an ordinary layered neural network, which has more hidden units (i.e. more parameters) to have more flexibility but does not update symbol representations.

Journal

Keywords

Details 詳細情報について

Report a problem

Back to top