Mapping recreational mobility based on crowdsourced data with the integratıon of spatial analysis methods and machine learning: The case of Eskişehir

The monitoring of recreational mobility is an important aspect in the management of sustainable tourism destinations. Current information and communication technologies provide innovative ways to collect data on recreational mobility. Location-based social media platforms provide large datasets to analyze the movements and behaviors of humans. In this study, it is aimed to modelling and mapping recreational mobility in Eskişehir destination with spatial-temporal analysis methods and machine learning integration based on crowdsourced social media data. The study includes applications of obtaining geotagged data from Foursquare, Flickr and Twitter platforms, investigating the correlation between the obtained data and official visitor data, classification of users as tourist and citizen, temporal analysis, density analysis, extraction of urban tourism interests using machine learning technique, creation of spatio-temporal flow maps based on complex network theory and space time cube based trend analysis and hotspot analysis. The results of the study reveal the importance and usage potential of geotagged social media data in analyzing the spatio-temporal dynamics of recreational activities. The study is expected to provide important benefits to all stakeholders of tourism industry, researchers and managers interested in marketing and planning tourism destination to ensure sustainable development of cities.