Research On Construction Of Mongolian Compound Noun Semantic Network

The Mongolian compound noun semantic network is a lexicalsemantic knowledge base built by using the Mongolian near synonyms compound noun sets as the concept representation and connecting multiple semantic relations and distinguishing features between concepts.At present,Mongolian information processing research based on grammatical information has made some achievements in lexical analysis and shallow parsing.In order to make computers capable of processing Mongolian information at a deeper level,the need to establish a Mongolian semantic knowledge base oriented to information processing is becoming more and more urgent.On the premise of systematic analysis of English and Chinese lexical semantic network and Mongolian lexicography,this paper selects more than 5,700 Mongolian compound nouns from Mongolian compound lexical database as the research object,and uses the theories and research methods of semantic field theory,literature method,sememe analysis and merge method to carry out lexical semantic analysis and applied research.This paper focuses on the academic issues related to the semantic network framework of Mongolian compound nouns,the concept database of Mongolian compound nouns,the concept relational database and the retrieval application platform based on it,and constructs the semantic network of Mongolian compound nouns reflecting the semantic features of Mongolian entity compound nouns.Its purpose is to apply it to lexicalsemantic analysis,information retrieval,word sense disambiguation,text proofreading,machine translation,knowledge reasoning and other application systems,and to provide a powerful and applicable lexicalsemantic knowledge base resource for computers to better understand and process Mongolian.This paper consists of four parts: introduction,five chapters,conclusion and references.In the introduction part,it expounds the reason of choosing the topic,the research status of lexical semantic network at home and abroad,the source of corpus,the research content and purpose,the research methods and research significance.Chapter One: In-depth analysis of theoretical research related to lexical semantic network.This chapter systematically discusses many research contents of Word Net,such as part of speech classification,concept representation methods,various relationships between concepts,database,specific implementation methods,and the overall situation of Mongolian dictionary compilation and research.The above analysis provides a theoretical and practical basis for the construction of the semantic network framework of Mongolian compound nouns.Chapter Two: Developing the framework of semantic network of Mongolian compound nouns.On the basis of referring to the research of English and Chinese lexical semantic network and Mongolian lexicography,according to the unique attributes of Mongolian and the present situation and needs of Mongolian information processing research,this chapter focuses on discussing and depicting the framework structure of Mongolian compound noun semantic network in two research parts: "Basic research of Mongolian compound noun semantic network" and "Concrete realization of Mongolian compound noun semantic network".Chapter Three: The conceptual database of Mongolian compound nouns is developed using Mongolian near synonyms compound noun sets as the conceptual representation.Firstly,based on the research of Mongolian near synonyms compound nouns,the classification system of three Mongolian near synonyms compound nouns,such as synonyms,near synonyms and semantic correlation,is put forward.Secondly,the definition,judgment method,semantic features,formal relations,composition,association and difference among various categories of Mongolian near synonyms compound nouns are deeply studied.Finally,a total of 4,905 Mongolian near synonyms compound noun sets composed of 6,406 words were established by means of manual work and human-computer interaction,thus constructing the concept database of Mongolian compound nouns.Chapter Four: The research on the construction of conceptual relational database of Mongolian compound nouns.On the basis of research on a number of theoretical issues,such as the definition,judgment method,semantic features,types,formal relations,and the association and difference among various relations of three semantic relations among Mongolian compound nouns,such as antonym,hyponym,and paranymy,as well as a Meronymy distinguishing feature,the above-mentioned multicategory relation databases among Mongolian compound nouns concepts are established respectively.At the same time,the method of establishing two distinguishing features of Mongolian compound noun concept,such as attributes and modification,functions and predication,is put forward and experimental research is made.Chapter Five: Design and implement the semantic network application platform of Mongolian compound noun.This chapter relates the data of Mongolian compound noun concept database and concept relation database,provides a semantic network database of Mongolian compound nouns with continuous expansion and modification functions for lexicographers,and also builds a semantic network retrieval platform for users.Thus realizing the retrieval of various semantic relations and distinguishing features between Mongolian compound noun concepts,and then completing the initial construction of semantic network of Mongolian compound noun.The conclusion part summarizes four main research contents:Mongolian compound noun semantic network framework,Mongolian compound noun concept database,Mongolian compound noun concept relational database and Mongolian compound noun semantic network application platform.At the same time,based on the research experience of this paper,the future research prospect is put forward.