A method of similary retrieval by using knowledge base of word relations. To classificate "100 poems by 10 poets".

DOI Web Site Web Site 1 Citations 2 References

Bibliographic Information

Other Title
  • 言葉の関連性による文書の類似検索　　小倉百人一首を分類・分析する
  • コトバ ノ カンレンセイニヨルブンショ ノ ルイジ ケンサク オグラ ヒャクニ
  • -To classificate"100 poems by 100 poets"-
  • ―小倉百人一首を分類・分析する―

Search this article

Description

Calculating similarity by frequency and interrelation of words is an ordinary method of similary retrieval or concept retrieval. In this paper, we use 4 methods of similary retrieval, similary measure, proximate co-occurrence, TF/IDF, and precision and similarity of documents, to classificate"100 poems by 100 poets". In similary retrieval, like knowledge retrieval by synonyms or thesauruses, it is nessesary to construct a reliable knowledge dictionary or concept dictionary automatically. Precision and similarity change according to the viewpoint of the user, so it is important to explain the process of the retrieval.

Journal

Citations (1)*help

See more

References(2)*help

See more

Keywords

Details 詳細情報について

Report a problem

Back to top