N元模型在藏文文本局部查错中的应用研究  

Research on the Application of the N-Gram Model in Tibetan Text Error Correction Partially by N-Gram

作　　者：多杰卓玛[1] 

机构地区：[1]西北民族大学中国民族信息技术研究院,甘肃兰州730030

出　　处：《计算机工程与科学》2009年第4期117-119,123,共4页Computer Engineering & Science

基　　金：甘肃省科技攻关项目(2GS064-A52-035-04)

摘　　要：本文阐述了藏文文本的错误形式的多样性。要对藏文文本进行局部查错,先要对文本中的音节进行纠错,音节的纠错又离不开对音节组成成分的纠错。本文利用N元模型对藏文音节内的各组成成分建立模型,并用此模型来判断一个音节的对与错。This paper describes the varieties of errors in Tibetan texts. To correct the partial errors, we need to correct the syllables firstly, which are made up of different parts. So it is important to correct the compositions of syllables. Based on these compositions of syllables, this paper establishes a model for error- correcting Tibetan texts. Therefore, the model can estimate if a syllable is right or not.

关 键 词：N元模型 藏文音节 字丁 同现概率 

分 类 号：TP391[自动化与计算机技术—计算机应用技术]