Abstract

This study explored the connections between behavioral and substance addictions in digital environments by identifying keywords that mediate links between these addictions on the social networking service platform X (Twitter). Posts were collected using representative keywords related to gaming, gambling, pornography, alcohol use, and drugs. After preprocessing, a total of 8,712 posts containing 5,027 unique keywords were analyzed. A semantic network was created based on a co-occurrence threshold of 200, employing Girvan–Newman community detection and edge betweenness centrality. The results revealed that platform- and function-based keywords, such as "RT," "Telegram," and "Line," frequently appeared across various addiction categories. In contrast, transaction- and participation-related terms (e.g., "inquiry," "sale," "purchase," "event") served as mid-level bridging keywords that enhanced connectivity between addictions. Additionally, even identical keywords showed different co-occurrence patterns depending on the addiction category. These findings indicate that the digital environment can facilitate mediation and transitions between addictions, blurring the lines between behavioral and substance addictions. They underscore the necessity for prevention and regulatory strategies to monitor and address the digital pathways that enable cross-addiction linkages.