텍스트마이닝을 활용한 청소년의 에너지드링크에 대한 인식 분석

Abstract

Objectives The purpose of this study is to explore the adolescents' perception of energy-drinks using big data. Methods In order to analyze big data, ‘adolescents’ and ‘energy-drinks’ related documents that had been generated from March 31, 2023 to March 30, 2024 were collected from major portal sites (“Naver”, “Daum” and “Google”) using Textom. A total of 15,959 documents were collected. Non-relevant and inaccurate words were removed for text mining. Among them, 50 words with a high frequency of appearance were extracted, and 50 keywords were derived around words with high meaning through refining. Results After that, frequency analysis, TF-IDF (Term Frequency - Inverse Document Frequency) analysis, centrality analysis, and CONCOR (Convergence of iteration Correlation) analysis were performed using NetDraw and the UCINET program, and clustering results between major keywords were investigated. As a result, Group 1 was routine caffeine intakes, Group 2 was purchasing facilitating factors, Group 3 was purchasing determining factors, and Group 4 was concerns of nutrient deficiency. Conclusions Based on these results, adolescents’ perceptions and issues of energy-drinks were identified. Furthermore, this study will provide theoretical implications for future research as well as practical implications for preparing policy regulatory measures and developing educational programs.