한국어 구어와 문어 말뭉치에 나타난 실질형태소의 출현빈도 상관관계 연구 -단순선형회귀 분석 모형을 중심으로-

Abstract

This paper introduced a simple linear regression analysis model, which is widely used in fields such as economic prediction, medical examination, civil engineering architecture, machine manufacturing, and machine learning to linguistics research, in order to conduct comparative studies on Korean spoken and written languages. This paper selected four full morphemes, which is general nouns, verbs, general adverbs and exclamations, took the frequency of occurrence of these full morphemes as a dependent variable, and the frequency of occurrence of other full morphemes as an independent variable to examine the relationship between them. First, it checked whether there is a linear correlation by drawing a scatterplot, and then the degree of correlation was examined by calculating the correlation coefficients A simple linear regression equation was constructed for independent and dependent variables with high correlation, and then a straight line was drawn to compare and analyze the differences between spoken and written words through the upper and lower relationships of straight lines and the size of narrow angles of the lines.