Construction And Quantitative Analysis Of The Knowledge Base Of Function Words In The Five Dialects Of Jiangsu

With the continuous development of the construction of basic dialect resources,the computer-oriented dialect corpora and knowledge bases have been established.However,existing basic dialect resources pay more attention to phonetics and content words.Function words,a main grammatical method of Chinese,are rarely involved.In the field of Chinese information processing,the information of function words plays an important role in automatically analyzing syntax and semantics of sentences.As to automatic analysis of dialects,it is also essential to utilize the grammar knowledge of dialect function words and establish the formal expressions for them.At present,the formal expression of function words in dialects has not been attempted,while the formal expression of function words in modern Chinese has been explored to some extent,but it is not perfect.Therefore,on the basis of combing,supplementing and utilizing the existing modern Chinese and dialect resources,this paper improves the formal expression system of function words in modern Chinese,and takes the knowledge of function words in five dialects of Jiangsu Province as an example to annotate and test,extract the formal expression system applicable to function words in Jiangsu dialects,And carry out relevant quantitative analysis.It makes up for the lack of function words in Jiangsu dialects in the analysis of appearance recognition and quantitative perspective.The specific research and results are as follows:(1)It improves the construction of the formal expression system of function words in modern Chinese.Based on the hypothesis that “the grammatical function and grammatical relationship of Modern Chinese function words can cover the information expression of dialect function words”,the integrity and comprehensiveness of Modern Chinese function words knowledge system is the foundation of constructing dialect knowledge system.Therefore,this paper refers to the formal and computable sememe labels in How Net,selects the Dictionary of Modern Chinese Function Words Usage for labeling,and annotates 559 words,covering 786 meanings.Comparing the two,it is found that there are still some problems in How Net.For example,function words in How Net are insufficient,and the number and type of sememe labels cannot cover the grammatical function of Chinese function words.So this paper combs the inconsistencies between the two based on the comparison and analysis of annotation results,and adds 37 more formal expression labels for those words that cannot be annotated by original How Net sememe labels,covering 89 function words expression labels,so as to realize the construction of a complete and standardized formal expression system of function words in modern Chinese.(2)The knowledge of function words in five dialect districts of Jiangsu Province is annotated and tested based on the mature formal expression system of function words in modern Chinese,and the formal expression system suitable for function words in Jiangsu dialect is attempted to be extracted.This paper selects the dialect dictionaries of Nanjing,Suzhou,Xuzhou,Yancheng,Yangzhou to annotate the function words information,considering the internal relations of three dialect areas of Jianghuai Modern Chinese,Wu dialect,Central Plains Modern Chinese in Jiangsu Province.In the process of annotation,this paper analyzes the problems of incomplete parts of speech and disunity in dialect dictionaries,and annotates the source corpus and the separation of senses when filling in entries.A knowledge base of function words in five dialect districts of Jiangsu Province with 437 words and 711 grammatical meanings was constructed,among the 57 function words in Nanjing,44 labels were used to annotate 104 meaning items;146 function words in Suzhou and 52 labels annotated 242 meaning items;66 function words in Xuzhou and 48 labels annotated 117 meaning words;92 function words in Yancheng and 49 labels annotated 134 meaning items;76 function words in Yangzhou and 48 labels annotated 114 meaning items.(3)Aiming at the problems existing in the process of labeling the knowledge base,the questionnaire survey was used to verify it.According to statistics in the knowledge base,it is found that the word"?"in the auxiliary dictionary of Xuzhou diale ct has the grammatical meaning of"do-first"used in the end of a sentence representing time,which cannot be expressed by the formal label of modern Chinese function words,so"Vdo-first|??"label is added.This illustrates the necessity of formal expression in the system of dialect words.There are the following problems in knowledge base labeling:1.Whether the function of the function word also exists in other dialect points is worth exploring;2.The function of the function word in some dialects is difficult to judge when labeling;3.Some formal labeling of Modern Chinese function words in the five local dialect function words in Jiangsu do not appear in the preliminary annotation of the knowledge base.Statistics found that only 30.12%of the formal tags appeared all in the four types of function words in the five local dialects of Jiangsu.In response to the above problems,the questionnaire survey was conducted to find out:1.In the other four dialect points in Jiangsu,the auxiliary word of Yancheng dialect also has the meaning of“do-first”;2.it is found that 96.53%of the labels are involved in the expression of function words in the dialect,which are limited by the scale of function words and example sentences in the knowledge base.However,there are still some individual formal tags that have not been investigated,such as the"{Vgoingon|??}"in the auxiliary system of Nanjing dialect,which provides a reference for showing the unique grammatical individuality of function words in each dialect district.In addition,the dialect words and example sentence of the five dialect points added in the questionnaire survey are explained,which enriches the words collection in the dialect knowledge base of dialect words.It can be seen that,except for the grammatical meaning of "do-first",the formal expression tags of modern Chinese function words can represent the function of the five local dialect words.It can be seen that the formal expression system of modern Chinese function words has certain adaptability to the system of function words in the dialect.It verifies the feasibility of constructing a formal expression system of function words in modern Chinese to express dialects,and provides a reference for the structural expression oriented to dialect information processing.(4)Through the verification and supplement of knowledge-based labeling and questionnaire survey,a quantitative analysis of the formal system of the functional words in the five local dialects of Jiangsu province was carried out.Compared with mandarin function words,the paper studied mainly from four aspects: same word form with same formal expression,different word form with same formal expression,same word form with different formal expression,same word form with same or different formal expression.In addition,this paper summarized the distribution of the formal words in the five local dialects of Jiangsu province in terms of consistency,inconsistency,similarities and differences.It is found that the dialect and tone words of all parties in the five regions of Jiangsu province are quite different from that of modern Chinese,and the average proportion of the tone words that are completely consistent with the modern Chinese words is only 33.32%.In addition,the internal disparity in the conjunction system of the four types of function word systems in the five local dialects of Jiangsu province shows great differences: Conjunctions in Nanjing dialect and modern Chinese account for 91.67%,which is basically the same as in modern Chinese;while in Suzhou dialect,conjunctions and modern Chinese are only 25%,there is a big difference in the use of conjunctions with modern Chinese.On the other hand,based on the comparision of modern Chinese,this paper further studied the cross-dialect points.Statistics of the three dialects in Nanjing,Yancheng and Yangzhou,both of which belong to the Jianghuai mandarin area,have a common modality function of 16.67%.The statics shows little consistent in the three areas.It can be concluded that the three dialects have distinctive features in terms of mood compared with modern Chinese.The five local dialects of Jiangsu province,which belong to the three major dialect areas,are distinctive in terms of conjunctions,prepositions and mood words compared with modern Chinese.In addition,this paper objectively sorted out the number of functional words and the distribution of formal expressions of the dialects of the five places in Jiangsu province.The average proportion of functional words in dialects with two or more functions was 30.21%.From the perspective of formalization,the functional differences of the common words in the three places in the Jianghuai Mandarin area were analyzed.Through the measurement and analysis of the formal expression of the dialect function words,the consistency and difference between the function words in the five local dialects of Jiangsu province,modern Chinese and the internal dialect points are compared to fully display the function word system in the five local dialects of Jiangsu province.