西里尔蒙古文同形词知识库的构建

发布时间：2018-05-25 07:54

本文选题：西里尔蒙古文 + 同形词　；参考：《内蒙古大学》2017年硕士论文

【摘要】：知识库资源开发是一项语言信息处理的基础性工作。随着语言信息处理技术的发展,知识库资源建设也需要同时开发。传统蒙古文知识库研究已经到达词法和语义等领域,传统蒙古文知识库资源丰富,但是西里尔蒙古文的知识库资源却十分稀缺。本文从《蒙古语词典》、《蒙汉词典》(增订本)、《新蒙汉词典》、及《蒙古语详解词典》等四部词典中选出同形词的词条,构建了"西里尔蒙古文的同形词信息词典"。并根据该词典构建了 "基于词典的西里尔蒙古文搭配词库"和"基于语料库的西里尔蒙古文搭配词库"。我们构建的"西里尔蒙古文的同形词知识库"由"西里尔蒙古文的同形词信息词典"、"基于词典的西里尔蒙古文搭配词库"和"基于语料库的西里尔蒙古文搭配词库"等三部分组成。本文由导论、第一章、第二章和第三章和结论等五个部分组成。导论部分主要介绍了研究目的,研究概况和研究方法。第一章,从语言信息处理的角度论述了西里尔蒙古文同形词的定义,分类以及同形词与多义词和兼类词的关系。第二章,阐述了"西里尔蒙古文的同形词信息词典"的构建过程,其中包括词条来源,选词原则及建库时遇到的困难和解决方法。最后介绍了"西里尔蒙古文的同形词信息词典"属性字段及取值规格。第三章,阐述了构建同形词搭配库的方法和步骤。同形词搭配库由"基于词典的西里尔蒙古文搭配词库"和"基于语料库的西里尔蒙古文搭配词库"等两部分组成。结论部分,总结了构建西里尔蒙古文知识库的过程中,遇到的困难和解决方法。同时提出了有待解决的问题和下一步的研究任务。
[Abstract]:Knowledge base resource development is a basic work of language information processing. With the development of language information processing technology, the construction of knowledge base resources also needs to be developed at the same time. The traditional Mongolian knowledge base has reached the lexical and semantic fields. The traditional Mongolian knowledge base is rich in resources, but the Cyrillic Mongolian knowledge base resources are very scarce. This paper selects homographs from four dictionaries such as Mongol Dictionary, Mongol Dictionary, New Mongol Chinese Dictionary, and Mongolian detailed explanation Dictionary, and constructs the Information Dictionary of homomorphic words in Cyrillic Mongolian. According to the dictionary, the lexicon of Cyrillic Mongolian collocation and Cyrillic Mongolian collocation based on corpus are constructed. The "knowledge base of homographs in Cyrillic Mongolian" is composed of "Information Dictionary of homographs in Cyrillic Mongolian", "Dictionary based on Cyrillic Mongolian collocation" and "Corpus-based Cyrillic Mongolian collocation". The library is composed of three parts. This paper consists of five parts: introduction, chapter one, chapter two and chapter three, and conclusion. The introduction part mainly introduces the research purpose, the research survey and the research method. The first chapter discusses the definition, classification and the relationship between homographs, polysemous words and concomitant words in Cyrillic Mongolian from the perspective of language information processing. In the second chapter, the author expounds the construction process of the Information Dictionary of homomorphic words in Cyrillic Mongolian, including the origin of the entries, the principles of word selection, the difficulties encountered in the construction of the database and the solutions. Finally, the attribute field and value specification of the Cyrillic Mongol homographs Information Dictionary are introduced. In the third chapter, the methods and steps of constructing homomorphic collocation database are described. The homographs collocation database consists of two parts: Cyrillic Mongolian collocation thesaurus based on dictionary and Cyrillic Mongolian collocation lexicon based on corpus. In conclusion, the paper summarizes the difficulties and solutions in the process of constructing the knowledge base of Cyrillic Mongolian. At the same time, the problems to be solved and the next research task are put forward.
【学位授予单位】：内蒙古大学
【学位级别】：硕士
【学位授予年份】：2017
【分类号】：H212

【参考文献】