Detecting Overtourism Signals through User-Generated Content: A Data-Driven Approach to Tourism Governance

Tourism drives economic and social development worldwide, yet urban destinations like Lisbon face challenges from overtourism, which can negatively affect cities. Traditional statistics capture flows and economic impacts but fail to reflect experiential and perceptual dimensions, limiting timely policy responses. This study leverages user-generated content (UGC) from 143,656 TripAdvisor reviews to detect signals of perceived overtourism. By integrating sentiment, thematic, and spatial analysis, the framework identifies activities experiencing visitor pressure, revealing a paradox of declining overtourism reports despite continued high visitation. Complaints concentrate on public transport and major cultural sites, while dispersed or smaller-scale activities show fewer concerns. The findings highlight a “reporting fatigue” as a perceptual adaptation to congestion and demonstrate the value of UGC for urban tourism management. Future work will incorporate spatial clustering and broader demographic data to enhance the framework’s applicability and guide sustainable destination governance.