摘要
利用日汉平行语料库自动获取汉语译词无论是在词典编撰,还是在自然语言处理领域都有着广泛的应用。以源词和译词之间的相互信息或Z值为依据,以译句对为评价范围,可有效地获取在只有单个译词情况下的汉语译词,并且采用这种方法可以得到较高的译词召回率和准确率。
Using JapaneseChinese parallel corpus to retrieve the Chinese equivalents automatically is generally applied not only in editing dictionaries but also in the field of natural language processing. On the basis of paired Japanese and Chinese sentences,according to the mutual information between the Japanese words and the Chinese equivalents or Z score,we can effectively get the Chinese equivalents. This approach can raise the rate of recall and precision.
出处
《解放军外国语学院学报》
北大核心
2003年第5期65-68,共4页
Journal of PLA University of Foreign Languages
关键词
日语词汇
汉语译词
自动获取
召回率
准确率
Japanese words
Chinese equivalents
automatic retrieval
rate of recall
rate of precision
