버토픽 기반의 토픽모델링을 활용한 청년고용 관련 텍스트의 다차원 분석 = Multidimensional Analysis of Youth Employment-Related Texts Using BERTopic-Based Topic Modeling

다국어 초록 (Multilingual Abstract)

Purpose The purpose of this study is to examine the differences between social and individual perceptions of youth unemployment and employment in Korea. By focusing on both public discourse and personal experiences, this study aims to highlight the emotional responses of young people that are not fully reflected in official statistics and to identify the perceptual gap that exists between society and individuals.


Design/methodology/approach This study collected and analyzed unstructured text data from news articles and YouTube comments covering the period of 2016 to 2025. Using BERTopic-based topic modeling, the analysis was divided into three distinct time periods. News articles were evaluated for their institutional and policy-oriented perspectives, while YouTube comments were assessed for emotional and personal reactions related to youth employment issues.


Findings The results revealed that news articles primarily emphasized institutional approaches and policy expansion, whereas YouTube comments reflected strong emotional responses such as anxiety, anger, and resignation. In the discourse on unemployment, conflicts between political tensions, individual responsibility, and structural causes were evident. In contrast, employment-related discourse focused more on practical and informational aspects. These findings confirm the existence of a significant gap between societal narratives and individual experiences, suggesting that future policy design should incorporate both structural solutions and emotional realities of youth.

다국어 초록 (Multilingual Abstract)

Purpose The purpose of this study is to examine the differences between social and individual perceptions of youth unemployment and employment in Korea. By focusing on both public discourse and personal experiences, this study aims to highlight the emotional responses of young people that are not fully reflected in official statistics and to identify the perceptual gap that exists between society and individuals.




Design/methodology/approach This study collected and analyzed unstructured text data from news articles and YouTube comments covering the period of 2016 to 2025. Using BERTopic-based topic modeling, the analysis was divided into three distinct time periods. News articles were evaluated for their institutional and policy-oriented perspectives, while YouTube comments were assessed for emotional and personal reactions related to youth employment issues.




Findings The results revealed that news articles primarily emphasized institutional approaches and policy expansion, whereas YouTube comments reflected strong emotional responses such as anxiety, anger, and resignation. In the discourse on unemployment, conflicts between political tensions, individual responsibility, and structural causes were evident. In contrast, employment-related discourse focused more on practical and informational aspects. These findings confirm the existence of a significant gap between societal narratives and individual experiences, suggesting that future policy design should incorporate both structural solutions and emotional realities of youth.