A Quantitative Study Of Mongolian Syntax Based On Legal Corpus

In recent years,there have been an endless stream of literatures using scientific quantitative linguistics to study stylistic features.Quantitative linguistics provides objective data support for the study of stylistic features,plays an important role in corroborating theoretical conclusions,and promotes the development of stylistic features quantitative research.This research attempts to use the "Civil Code of the People’s Republic of China"(Mongolian version)as the corpus,and establishes a dependency tree database,and extracts measurable indicators from the tree database,and attempts to answer the following three main questions with real data statistical analysis research:(1)What are the characteristics of the syntactic structure of Mongolian legal style?(2)What are the characteristics of the syntactic functions of verbs,nouns and adjectives in Mongolian legal style?(3)What are the characteristics of the Mongolian legal style dependent syntactic network?For the first research question,the dependency treebank of Mongolian legal style is firstly constructed,and then sentence length,sentence pattern distribution,sentence pattern distribution,dependency distance,dependency type characteristics and other quantifiable indicators of Mongolian legal style are extracted from 2864 sentence in the treebank,and correlation analysis is also carried out.At the same time,Sentence length includes average sentence length,sentence length dispersion,sentence length skewness,sentence length kurtosis and sentence length probability distribution.The research of dependency distance includes average dependency distance,dispersion degree of dependency distance and probability distribution of dependency distance.Based on the statistical analysis results of relevant data,the syntactic structure characteristics and rules of Mongolian legal style are analyzed and summarized.For the second research question,We first examine the situations in which various parts of speech act as dominant and subordinate words.Then according to these situations,the syntactic functions of nouns,verbs and adjectives are analyzed emphatically.On this basis,a comparative study with the syntactic function of the part of speech of the common language style is carried out.For the third research question,the construction method of language network and the processing of corpus are firstly introduced.On this basis,the word co-occurrence network and dependency syntactic network of Mongolian legal style are established.Secondly,the basic properties and complex network characteristics of the two language networks are analyzed.After comparative research,it is found that the dependency syntactic network can better reflect the characteristics of sentences.Therefore,in order to highlight the characteristics of Mongolian legal style dependent syntactic network,a Mongolian universal language dependent syntactic network was established,and a comparative analysis was conducted with it.This research builds a Mongolian legal style dependency treebank,expands the scope of Mongolian corpus and the field of quantitative research,provides a new perspective and technical route for legal style research,achieves innovative results,and promotes linguistics and Mongolian research.