基于编辑距离的语言分类研究  

Language Classification Study Based on Levenshtein Distance

作　　者：赵志靖[1] 江荻[2] ZHAO Zhi-jing;JIANG Di(School of Journalism and Communication,Yangzhou University,Yangzhou Jiangsu 225000,China;Institute of Ethnology and Anthropology,Chinese Academy of Social Sciences,Beijing 100081,China)

机构地区：[1]扬州大学新闻与传媒学院,江苏扬州225000 [2]中国社会科学院民族学与人类学研究所,北京100081

出　　处：《语言研究》2020年第2期43-50,共8页Studies in Language and Linguistics

基　　金：教育部人文社会科学研究青年基金项目“江苏方言语音相似关系的计量研究”(15YJC740141);江苏省社会科学基金一般项目“长三角地区汉语方言语音相似关系计量研究”(18YYB014)。

摘　　要：编辑距离被证明是研究西方语言之间语言距离的有效方法,但东亚语言的研究未有涉及。基于Almeida & Braun调音系统计算元辅音间距离,然后以此值修订编辑距离算法的"代价",计算语音之间的距离,可以得到语言之间的距离。同时,应用编辑距离算法对藏语6种方言进行距离计算,表明编辑距离的分类结果与历史语言学的分类结果是完全一致的。编辑距离可以应用于东亚语言的研究中。Levenshtein distance has proved to be effective to measure the language distances between western languages, whereas East Asian languages have seldom been involved. In this paper, vowel and consonant distances are calculated based on the Almeida&Braun articulation-based system. Sound distances are calculated by revising the operation cost of the Levenshtein distance algorithm, thus language distances are obtained. At the same time, this paper applies the Levenshtein distance algorithm to six Tibetan dialects. The calculation result shows that the language classification of the Levenshtein distance is consistent with that of the historical linguistics. The Levenshtein distance can be applied to the research of the East Asian languages.

关 键 词：编辑距离 Almeida&Braun调音系统 语言分类 

分 类 号：H146[语言文字—汉语]