藏文词同现网络的构建及特征分析  

Research on Construction and Feature Analysis of the Tibetan Words Co-occurrence Network

作　　者：加羊东周 才智杰[1,2,3,4] 才让卓玛 三毛措[1,3,4] Jiayang-Dongzhou;Caizhi-Jie;Cairang-Zhuoma;San mao-Cuo(College of Computer Science and Technology,Qinghai Normal University,Xining 810016,China;School of Computer Science and Technology,Southwest Minzu University,Chengdu 610041,China;Tibetan Information Processing and Machine Translation Key Laboratory of Qinghai Province,Xining 810008,China;Key Laboratory of Tibetan Information Processing,Ministry of Education,Xining 810008,China)

机构地区：[1]青海师范大学计算机学院,青海西宁810016 [2]西南民族大学计算机科学与技术学院,四川成都610041 [3]藏文信息处理教育部重点实验室,青海西宁810008 [4]青海省藏文信息处理与机器翻译重点实验室,青海西宁810008

出　　处：《高原科学研究》2020年第3期111-116,共6页Plateau Science Research

基　　金：国家自然科学基金项目(61866032,61966031);青海省科技厅资助项目(2019-SF-129);“长江学者和创新团队发展计划”创新团队资助项目(IRT1068);青海省重点实验室项目(2013-Z-Y17,2014-Z-Y32,2015-Z-Y03);藏文信息处理与机器翻译重点实验室(2013-Y-17)。

摘　　要：语言同现网络用复杂网络技术对语言进行研究,揭示语言的内部结构。文章收集和整理了6个不同载体的藏语语料,包括诗歌、散文、政治、佛教、教材及口语共计64篇文章,在此基础上构建了藏文词同现网络,并从最短路径、聚类系数、度分布等方面分析了藏文词同现网络的统计特征,实验表明藏文词同现网络具有小世界效应和无标度特性。Language co-occurrence network adopts various complex network technologies to study language toreveal the internal structure of language itself.Based on the collection and processing of 64 textual Tibetan docu-ments in six major categories,including poetry,prose,politics,Buddhism,textbooks and spoken language,the ar-ticle construct a Tibetan word co-occurrence network.The statistical characteristics of the Tibetan word co-oc-currence network are analyzed via Shortest Path,Clustering Coefficient,degree distribution etc.Experiments em-pirically indicates the Tibetan word co-occurrence network has a small-world effect and scale characteristics.

关 键 词：同现网络 藏文词 小世界效应 无标度特性 

分 类 号：TP391.1[自动化与计算机技术—计算机应用技术]