未分化な文法カテゴリによる幼児発話の誤用

DOI Web Site

書誌事項

タイトル別名
  • Errors in Children’s Speech Originating from Undifferentiated Grammatical Categories
  • 未分化な文法カテゴリによる幼児発話の誤用 : 英語の過去形の形態素と日本語の格助詞の過剰生成に共通した計算モデル
  • ミブンカ ナ ブンポウ カテゴリ ニ ヨル ヨウジ ハツワ ノ ゴヨウ : エイゴ ノ カコケイ ノ ケイタイソ ト ニホンゴ ノ カクジョシ ノ カジョウ セイセイ ニ キョウツウ シタ ケイサン モデル
  • 英語の過去形の形態素と日本語の格助詞の過剰生成に共通した計算モデル
  • A Model for Overproduction of an English Past-Tense Morpheme and a Japanese Case Particle

この論文をさがす

抄録

 Young children produce multi-word sentences including some systematic errors or<br> overproduction. It has been reported that English-speaking children may add a mor-<br>pheme “ed” to an irregular verb as its past tense while Japanese-speaking children may<br> position a case particle “NO” after an adjective. We hypothesize that an insufficient in-<br>crease in grammatical categories causes such overproduction, which can be expected to<br> disappear with a sufficient increase.We assume that hidden states of a hidden Markov<br> model (HMM) correspond to grammatical categories acquired from language input.<br> Based on the HMM, the simulation results could partially verify the above hypothesis.<br> In the English-trained model, the overproduction could appear and then decline. How-<br>ever, it did not completely disappear because categories of regular and irregular verbs<br> did not differentiate even when the model had many categories. In the Japanese-trained<br> model, the overproduction could appear and then disappear through differentiation of<br> categories of nouns and adjectives. The limitations of the proposed model are pointed<br> out and future issues are discussed.

収録刊行物

  • 認知科学

    認知科学 24 (1), 55-76, 2017

    日本認知科学会

関連プロジェクト

もっと見る

キーワード

詳細情報 詳細情報について

問題の指摘

ページトップへ