A Study of Modern Chinese Quantifier Constructions Based on Random Forest (RF) Modeling

This paper attempts to conduct a systematic study on the constructions of quantity phrases in modern Chinese on the basis of relevant research results, drawing on the theory of constructive grammar, in order to demonstrate the mechanism of constructions of quantity phrases in modern Chinese.The study firstly researches and analyzes the matching and distribution of quantity phrases as well as the Chinese construct grammar.Then, the study is based on random forests to investigate the constructions of quantifiers.By extracting and labeling six modern Chinese corpora, the analysis is carried out using random forests.On this basis, in order to further analyze the role of the relationship between the constructions of quantifiers, this paper also invokes a multinomial logit regression model for the study.It is found that the construct variant, regional variant, verb immediately following at the end of the sentence, structure initiation, and verb prototypicality are important factors affecting the number word constructions.In addition, the probability of quantifiers was higher when the construction variants were A and D, and sentence initiation while more inclined to co-occur with quantifiers.These findings reveal constraints on quantifier constructions and demonstrate the advantages of combining machine learning methods to analyze Chinese constructions.