國家教育研究院

　　本計畫探討各國應用語料庫於詞典編輯的技術，利用文獻分析及演算法實作等方法，自動從大量的語料中爬取出包括搭配詞、同義詞、反義詞、語法結構及詞彙速寫等中文語言知識，來輔助中文學習者詞典的編輯，提高詞典編輯的效率、客觀性及實用性。

　　This project aims to apply the natural language processing technology on Chinese learner’s dictionary compilation. We adopt the approaches of literature analysis and different algorithms to acquire word information like synonyms, antonyms, word sketches and grammatical patterns from corpus automatically, to improve the efficiency, objectivity and practicality of dictionary compilation.