This year, the 5^th edition of the prestigious International Data Science Conference (iDSC) was held from May 2-3, 2023, and we received an invitation to attend and present our paper on “Prediction of Tourism Flow with Sparse Geolocation Data”. Tourism in the 21^st century puts all aspects of human life to the test. In particular, the harmonious coexistence between locals and tourists as well as the spatial capacities of tourist attractions are overloaded. Managing these masses and coordinating a pleasant coexistence is a challenge. In addition, the COVID-19 pandemic has also taught us that we must be flexible to ensure a stable environment even in times of crisis. For this reason, forward-thinking tourism regions first strive to understand and then control visitor flows to preserve and protect their natural landmarks. By predicting tourist flows and volumes, we can adapt to different situations and adjust our actions accordingly. This significantly increases effectiveness and prevents unforeseen overcrowding of attractions. This is the focus on an ongoing project with the City of Salzburg, which we presented at this conference. By evaluating several different prediction models, we aim to create a way to manage tourist crowds in a way that does not overcrowd attractions and optimizes the experience for visitors and locals alike. To increase the accuracy of these predictions, we used advanced input attribute processing methods by mapping geolocation data to discrete points of interest (POIs).

More about this project can be found in our blog post : Tourism Flow Guidance Using Salzburg City as an Example https://datenvorsprung.at/2023/03/09/elementor-1824/

Link to our paper: https://arxiv.org/abs/2308.14516