A Quantitative Study Of Syntactic Features Of Mongolian Nouns Based On Treebank

Reviewing the linguistic research in Mongolian,it is found that most of them mainly use qualitative methods to describe language,and only a few use quantitative methods.In this paper,we use the methods of quantitative linguistics to explore the academic issues related to nouns and noun phrases in Mongolian based on dependency grammar and phrase structure grammar in written language and spoken language treebanks respectively.The paper consists of six chapters: introduction,corpus and methods,quantitative study of syntactic properties and functions of Mongolian nouns,quantitative study of syntactic properties and functions of Mongolian noun phrases,comparative study of Mongolian noun syntactic features based on treebanks,and conclusion.The main contents are as follows:The first chapter mainly introduces the reasons of topic,research status,research significance and research content.The second chapter introduces the sources of written and spoken language corpora,corpus processing steps,the main research methods and data analysis tool.The third chapter studies the quantitative characteristics of Mongolian noun distribution,the dependence distance distribution and dependence direction distribution of nouns as governors and dependents in written and spoken treebanks,as well as the syntactic valency analysis and related part of speech distribution of nouns in written and spoken treebanks.The fourth chapter analyzes the quantitative characteristics of Mongolian noun phrase distribution,the PIT frequency distribution,depth distribution,length distribution of noun phrases based on written and spoken treebanks,as well as the syntactic function distribution and related part of speech sequence distribution of noun phrases based on written and spoken treebanks.The fifth chapter discusses the comparative study of dependence distance and dependence direction of nouns based on written language and spoken language treebanks,the comparative study of PIT frequency,depth and length of noun phrases based on written language and spoken language treebanks,and the comparative study of syntactic functions and related parts of speech of nouns and noun phrases based on written language and spoken language treebanks.The sixth chapter summarizes the research results,innovations,research gaps and future research.The results show that in terms of dependency distance,in written and spoken language,whether nouns are used as governors or dependents,their dependency distance distribution conforms to the right truncated Waring distribution and right truncated modified Zipf-Alekseev distribution.In terms of dependency direction,whether in written or spoken language,whether nouns are governors or dependents,the results show that tend to the head-final type.In terms of the frequency distribution of the pattern of immediate constituents of noun phrases,it fits well with right truncated modified Zipf-Alekseev distribution both in the written and spoken treebank.In terms of noun phrase depth,it is found that the noun phrase depth data in written and spoken language treebanks fit well with the negative binomial distribution.In terms of noun phrase length,it is found that the noun phrase length in the written language treebank fits well with the mixed negative binomial distribution.The noun phrase length in spoken language treebank fits well with the extended logarithmic distribution.In terms of syntactic function,whether nouns are used as governors or dependents,whether in written or spoken language,the attribute accounts for the largest proportion of all dependency types.From the perspective of the proportion of noun phrase structural relationship,the attribute ranks first in both written and spoken languages.From the perspective of the internal structure of noun phrases,noun phrases in written and spoken language are mainly composed of non-embedded and embedded noun phrases with attribute relationship.From the perspective of the external syntactic function of noun phrases,noun phrases in written and spoken language mainly form attribute,subject,object,auxiliary and coordinate with the upper structure.Compared with traditional linguistic research,this paper has a certain novelty in terms of research methods which promotes the research of quantitative linguistics in Mongolian academia.It also further expands the existing research on quantitative linguistics.The results of this paper provide practical support for language teaching and application development of natural language processing.In the follow-up study,we plan to further research the syntactic features of other parts of speech in the dependency treebank and phrase structure treebank,research the similarities and differences of other parts of speech in written and spoken languages.In addition,we will also study the syntactic attributes and syntactic functional features of nouns and noun phrases in other languages,and then explore whether there are the same universal laws as the syntactic features of nouns and noun phrases in Mongolian.We intend to further expand and deepen basic linguistic research and application development.