基于语料库的“感动”一词词类标注问题研究

Corpus-based Study on Part-of-speech Tagging for “gandong”

汉语兼类词一直是一个老大难的问题,兼类词的词性标注更是语文辞书编纂的难点。文章以语料库为研究工具,对现代汉语高频使用的"感动"的词性进行了查阅和探讨,发现目前大中型汉语词典对"感动"的词性处理有失允当,因此,在利用语料库占有充分语言材料的基础上,对"感动"的词性进行了深入探讨。

Chinese multi-category words have always been a daunting problem. Furthermore, word class tagging is the difficulty in compiling the Chinese dictionaries. With corpus as research instrument,“gandong”,a frequently used Chinese word,is analyzed. It is discovered that there are many cases of mis-interpretation on word category in big and medium sized dictionaries. The word class of “gandong” is discussed on the basis of corpus that contain full language material.