A Preliminary Study On Automatic Recognition Of Improper Chinese Collocation

The purpose of this paper is to make use of computer to recognize speech disorders.However,the research on automatic proofreading of Chinese text has been a difficult problem to be solved since the beginning.In addition,the fact of Chinese language is complex,with different levels and categories of linguistic diseases,so the research on automatic recognition of textual linguistic diseases needs to be carried out at different levels.Referring to the research results of previous scholars on the types of Chinese language disorders,this paper studies inappropriate Chinese collocation language disorders from the perspective of feasibility and necessity.The research on automatic recognition of inappropriate collocations is carried out from the following parts: The first part explains the reason of topic selection and research significance;Summarize the research status and development trend related to this paper;The main work and structure arrangement of this paper are briefly described.This paper introduces the work related to this paper,namely the definition and classification of lexical diseases,corpus construction,dictionary resources.The second part determines the evaluation criteria of collocation language diseases.According to the method of description of composition collocation,language diseases are divided into three categories: inappropriate collocation,inappropriate use of verbs,inappropriate use of subjects,inappropriate use of additional elements and inappropriate use of central language.This paper focuses on describing the characteristics of all kinds of linguistic diseases and analyzing the causes of linguistic diseases from the aspects of syntax,semantic relations and collocations.In the third part,based on the research in the second part,a knowledge base of inappropriate collocations is established,which mainly introduces the concept of language knowledge base,gives an overview of the content of the knowledge base of inappropriate collocations,displays the knowledge base of various minor collocations,and analyzes the important data in the knowledge base.In the fourth part,the author selects some typical examples of inappropriate collocation to carry out further research,including collocation dictionary and language knowledge.To make themformal description,find out the rules,provide more knowledge for computer to recognize language diseases;Imagine the identification process,discuss the possibility of realizing the identification,and explain the difficulties encountered at present.The fifth part summarizes the whole paper,puts forward the existing problems and the future prospect of the subject.