LDA 토픽모델링과 의미연결망 분석을 통한 국내 수소산업 동향 분석

초록

Hydrogen has been gaining attention as an eco-friendly energy source that can replace fossil fuels in addressing the climate crisis. In South Korea, following the government's announcement of the <Hydrogen Economy Roadmap> in 2019, various policies have been established and implemented to foster the hydrogen industry. Correspondingly, a significant number of news articles related to the hydrogen industry have been reported. This study aims to identify industry trends by conducting a text mining analysis of news articles related to the domestic hydrogen industry. For this purpose, approximately 13,000 news articles published between 2019 and 2024 are analyzed using text analysis techniques such as Latent Dirichlet Allocation (LDA) topic modeling and semantic network analysis(SNA). Through this approach, both the overall trends in the news articles and the changes in trends over time are examined. The analysis reveals the presence of nine distinct topics, which collectively describe the trends and prospects of the hydrogen industry in South Korea. Among these, the installation of hydrogen refueling stations and the expansion of hydrogen vehicle adoption are identified as the most prominent subjects based on their topic proportions. Additionally, SNA explores the relationships between key terms, and the annual changes in topic proportions provided insights into the evolving trends in news coverage. This study holds significance in that it identifies the primary topics of interest in domestic news articles regarding the hydrogen industry through topic modeling and semantic network analysis. Furthermore, this research contributes practical implications for guiding the growth and development of the hydrogen sector.

키워드