WANDERPREDICT: UNDERSTANDING TOURIST PATTERNS THROUGH SOCIAL MEDIA DATA

Abstract

Tourism behavior analysis has become increasingly data-driven with the rise of social media platforms that capture valuable information about travelers’ preferences, interests, and mobility patterns. This study introduces a predictive framework that leverages social media user profiles to forecast tourist movements and travel trajectories. By applying machine learning and data mining techniques to geotagged posts, check-ins, and user-generated content, the system identifies mobility patterns, travel hotspots, and seasonal trends. The proposed approach enhances destination management, resource allocation, and personalized travel recommendations by offering real-time insights into tourist behavior. Experimental results demonstrate the effectiveness of social media–based prediction models in improving the accuracy of mobility forecasts compared to traditional survey-based methods. This research highlights the potential of integrating social media analytics into smart tourism systems, paving the way for data-driven decision-making in tourism planning and sustainable destination management.