藏语语料库加工方法研究  

Tibetan corpus processing method

作　　者：才让加[1] 

机构地区：[1]青海师范大学藏文信息研究中心,西宁810008

出　　处：《计算机工程与应用》2011年第6期138-139,146,共3页Computer Engineering and Applications

基　　金：国家语委基金(No.MZ115-018);国家社会科学基金(No.07BYY035);国家社会科学重点基金(No.05AYY001);973计划前期研究专项(No.2010CB334708)

摘　　要：为了使藏语语料库具有规范性、统一性和实用性,提高加工的整体水平,在藏语语料库的加工过程中首先要对五花八门的藏语语料库进行整理和统一,得到高质量的原始语料库,其次确定藏语原料库加工的切分单位,针对藏语的语法特征提出藏语语料库藏语词语类别和词类标记集,同时在对藏语词语进行归类和统计的基础上建立分词标注词典库,设计并实现藏文自动分词标注软件,利用分词标注软件对大规模藏语语料库进行切分和标注,最终实现藏语语料库的多级加工。In order to make the Tibetan corpus standardization,unity,practicability and to improve the overall level of pro- cessing.The multifarious Tibetan corpus in the processing part should be arranged and unified,which can get high quality of raw corpora.Then the processing units of Tibetan ancillary facilities for segmentation is determined,the Tibetan language syntax category and Tibetan words corpus are put forward, and participles tagging dictionary is set up based on the selection in the words of Tibetan categorize and statistics.The Tibetan automatic word segmentation tagging is designed and carried.The large-scale Tibetan corpus is segmented and labeled by using participle labeling software.The multilevel processing of Tibetan Corpus is implemented.

关 键 词：藏语语料库 规范 词类 标记集 词典 分词标注 

分 类 号：TP391[自动化与计算机技术—计算机应用技术]