LDA 기반 토픽 모델링을 활용한 소셜미디어 청년창업 외식 텍스트 분석 연구

This study explores youth entrepreneurship and the restaurant industry through social big dataanalysis, aiming to derive implications for youth entrepreneurship support policies. To this end, textdata related to youth entrepreneurship and the restaurant sector were collected from major portalsand social media platforms, including Naver, Daum, and Google, between April 1, 2022, and April1, 2024. The analysis was conducted using Python-based text processing tools such as Gensimand KoNLPy within Google Colab, leveraging cloud computing to enhance efficiency andreproducibility. Applying Latent Dirichlet Allocation (LDA) topic modeling, the study identified fourkey topics: restaurant entrepreneurship types, regional startup support, the food industry andinvestment, and youth entrepreneurship support and education. The findings indicate that whilejob postings related to youth entrepreneurship are actively shared, information on risks andoperational challenges is lacking, with an emphasis on benefits. The study also highlights thepotential for linking youth entrepreneurship with local communities, suggesting that regionalentrepreneurship support could contribute to economic revitalization. However, most initiativesremain short-term, lacking systematic policy integration. This study presents a novel approach toyouth entrepreneurship research through big data analysis and underscores the need for policiesthat better align with real-world entrepreneurial environments. Policy recommendations includestrengthening information provision in early-stage entrepreneurship, developing region-specificstartup models, and introducing sustainable support programs. The findings provide a foundationalresource for future youth entrepreneurship policy development and refinement.