Trend Investigation of Hairstyles Using Big Data Analysis Before and After of COVID-19: Focusing on Text Mining and Semantic Network Analysis

Abstract

Recently, hairstyle trends have changed drastically in relation to K-beauty. Therefore, this study performed an atypical big day analysis to analyze hairstyle trends before and after COVID-19. Next, Naver and Google collected a total of 65,776 keywords using Textom, a text data analysis program, from the last four years (two years from January 2018 to December 2019 before COVID-19 and two years from January 2020 to December 2021 after COVID-19). And by refining unnecessary words for text mining, 50 key keywords were derived each before and after COVID-19 and used for empirical analysis. In the top 50 keywords, it was found that hair, perm style, men, women, cuts, and short hair were the main keywords. In the semantic network analysis using UCINET 6.0 and Netdraw, it was confirmed that keywords such as hair, perm style, male and female, change, and cut are highly connected with other keywords. Finally, as a result of CONCOR analysis, before COVID-19, it was divided into four clusters: styling, cut style, promotion, and style transformation, and after COVID-19, it was divided into three clusters: style transformation, promotion, and styling. These analysis results confirm trends related to hairstyles, key components and promotional channels for hairstyle fields.