• 제목/요약/키워드: Large-scale N-gram Corpus

검색결과 1건 처리시간 0.013초

Web Catchphrase Improve System Employing Onomatopoeia and Large-Scale N-gram Corpus

  • Yamane, Hiroaki;Hagiwara, Masafumi
    • International Journal of Fuzzy Logic and Intelligent Systems
    • /
    • 제12권1호
    • /
    • pp.94-100
    • /
    • 2012
  • In this paper, we propose a system which improves text catchphrases on the web using onomatopoeia and the Japanese Google N-grams. Onomatopoeia is regarded as a fundamental tool in daily communication for people. The proposed system inserts an onomatopoetic word into plain text catchphrases. Being based on a large catchphrase encyclopedia, the proposed system evaluates each catchphrase's candidates considering the words, structure and usage of onomatopoeia. That is, candidates are selected whether they contain onomatopoeia and they use specific catchphrase grammatical structures. Subjective experiments show that inserted onomatopoeia is effective for making attractive catchphrases.