Research And Implementation Of Mongolian Coding Conversion Based On Rules And Statistics

With the rapid development of the computer technology and the network technology, Mongolian information processing have also made great progress.As Mongol(?)an international standardized coding is relative lagging,all the research units have respectively adopted their own Mongolian coding systems.All the above made different Mongolian data and web sites can not be compatible,information can not be shared,thereby seriously affecting the Mongolian Internet development.At present the majority of Mongolian information and Web site have adopted the Mongolian coding systems based on the word shapes.This paper mainly discusses the conversion from Menkeli Mongolian coding,Oyuta Mongolian coding,Saiyin Mongolian coding to the Mongolian international standardized coding.In order to achieve the conversion with unified approach,we use the Min-Morpheme coding to converse them to the Mongolian international standardized coding. Since the whole process of conversion is from the glyph coding to the sound coding,how to solve the different pronunciation of the same shape characters has become difficult problems to be solved in this article.The main work is divided into two parts:Firstly,we must daft the rule correspondence table of Mongolian glyph coding and the Min-Morpheme coding,and converting codes based on it. Secondly,three methods are used to achieve codes converting from the min-morpheme coding to the Mongolian international standardized coding.They are the method based on rules correspondence table,the method based on the Mongolian orthography dictionary,and the method based on statistical language model,and comprehensive use of the above measures to improve the conversion correct rate,and achieved the desired results.