공항산업 동향분석을 위한 텍스트 애널리틱스 모델에 관한 연구

Analyzing current trends and identifying future prospects is one of the important tasks to establish a successful strategy for airport development. In this study, we use latent Dirichlet allocation (LDA), a typical topic modeling technique, to supplement the traditional qualitative methodologies and find a major trend of airport industry. Specifically, we deduct major topic from three different data sources related to airport (academic article, policy study report and news article) and compare each topic to find out the differences of each topic. In addition, we investigate the changing trend in each data source through longitudinal analysis. Our results show that three data sources have distinguishing trends which reflect their own characteristics. We find that most of data usually focus on the future situation for sustainable development rather than concentrating on the current issues. This study suggests a methodology using large amounts of text data to explore future promising trend from sperate data sources and provide insights for stakeholders of airport industry. The analyzing methodology we proposed has also significant meaning in that it can be applied comprehensively for trend analysis of other industry including tourism and hospitality as well as airport industry.