摘要
藏文文本的自动校对是藏文信息处理的主要应用领域之一,现代藏文词的自动校对是其中重要的关键技术之一。本文根据藏文词错误的类型,分别提出了对应的校对方法。对非词错误,采用分词切分和散串匹配处理方法。对于真词错误,主要检查词的二元接续关系和词性邻接关系,均取得了较好的校对效果。
Tibetan text automatic proofreading is one of the main application fields of Tibetan information processing. Modern Ti-betan word automatic proofreading is one of the important key technologies. This paper makes proofreading methods based on different types of Tibetan word error. Non word error proofreading could be done by word segmentation and string matching pro-cessing. True word error proofreading could be done by checking the binary relationship of words connection and adjacency rela-tionship of words. Results of both proofreading methods are good.
作者
刘芳
关白
LIU Fang, GUAN Bai (Tibetan Information Technology Research Center in Tibet University, Lasa 850000, China)
出处
《电脑知识与技术》
2016年第7期200-201,203,共3页
Computer Knowledge and Technology
基金
西藏自治区自然科学基金资助项目“基于统计声学建模的藏语语音合成技术研究”(编号:2015ZR-14-12)
2012年度国家自然基金项目“现代藏文自动校对研究”(项目批准号:61202189)
关键词
藏文文本
自动校对
非词
真词
Tibetan Text
Automatic Proofreading
Non Word
True Word