Developing a text mining-based process for comparative analysis of trends in periodic trade shows: Focusing on CES

This study aims to analyze the key characteristics and trends of international exhibitions by examining news article text data from the Consumer Electronics Show between 2022 and 2024. While domestic companies across various industries recognize international exhibitions as significant strategic opportunities for global market expansion, the fragmented and inconsistent nature of exhibition-related information across multiple platforms poses a challenge in selecting the most suitable exhibitions and making informed strategic decisions. To address this issue, this study employs Latent Dirichlet Allocation topic modeling and sentiment analysis, two key techniques in natural language processing, to establish a standardized process for analyzing exhibition trends. The findings provide actionable insights that enable domestic companies to make data-driven strategic decisions, select exhibitions aligned with their goals, and develop effective global marketing strategies.