A Statistical Study On The Affix Construction Of Mongolian Verbs For Information Processing

Mongolian verbs are characterized by multiterminal morphologic changes and complex category changes.There are many academic achievements on the correct recognition and generation of Mongolian verbs in the research of Mongolian information processing.And get better results.However,it is still necessary to optimize and improve the rate of Mongolian verb recognition and accurate generation in information processing.In order to improve the recognition rate and the rate of automatic and accurate generation,the author makes a more comprehensive understanding of the combination relation between the conformation adjuncts of Mongolian verbs and the combination relation between the word root and conformation adjuncts.Based on the Mongolian ten-million-level corpus,this paper applies the methodology of quantitative linguistics and corpus linguistics to calculate 80137 unrepeated Mongolian verbs.Then,the root and conformation adjunctive components of these verbs are separated,and the relation library of root and conformation adjunctive components is established.In the library,the combination relation between the word root and various configuration adjuncts and the combination relation between each configuration adjuncts are embodied.The structure and contents of the paper are as follows: the introduction introduces the significance,purpose,source of corpus,research status and research steps of the topic.The first chapter describes the basis and principles of word selection,the process of corpus processing,the construction of relational database and the development of search system.The second chapter analyzes the general situation of Mongolian verbs and their conformal adjuncts,the combination of conformal adjuncts and the combination of Mongolian verb roots and conformal adjuncts.The third chapter introduces the practical significance of this research in practical application.The conclusion summarizes the whole paper.