以《尼山萨满》为例,利用语料库的技术手段保护和传承少数民族文化典籍.构建了一个满族典籍的平行语料库系统原型,重点研究了该平行语料库的语料对齐方法,分别研究了两种段落对齐方法和三种句子对齐方法,并对各个方法进行了性能评价,最终选出最适合该平行语料库的对齐方法.测试结果表明,最终选取的对齐方法满足构建满族典籍平行语料库的需求,从而为其他同类型的少数民族语料库构建提供更多的参考.%Taking Nishan Shaman as an example, we attempt to inherit and protect ethnic minor-ity cultural classics with the help of modern technology. A prototype of Manchu classics parallel corpus is constructed on the basis of Nishan Shaman. And this paper focuses on corpus align-ment of ethnic classics parallel corpus, where two kinds of paragraph alignment mothods and three kinds of sentence alignment methods are studied and functionally evaluated. The best a-lignment methods for this corpus are eventually selected. The assessment result shows that the a-lignment methods of this study are quite effective and efficient on constructing the Manchu clas-sics parallel corpus, which will provide a reference for the construction of similar ethnic classics parallel corpus.
展开▼