Core Projects - Mobility & Tourism

The study of human mobility can have a tremendous impact on several aspects of our society, such as disease spreading (i.e. COVID-19), urban planning, well-being, pollution, immigration, behavioural studies, etc. The proliferation of digital mobility data, such as phone records, GPS traces, and social media posts, combined with the unprecedented predictive power of artificial intelligence, triggered the application of deep learning to human mobility.

On the other hand, mobile devices have enabled the generation of vast amounts of data, but these data are typically owned by mobile phone network operators or mobile phone operating system developers (i.e., Google and Apple). Therefore, only a few datasets are publicly available; when it does, it often comes with geographical area, period, and population sample size limitations.

Therefore, to overcome the current limitations presented by the available datasets, we intend to define a methodology, dubbed GeoProMob, that can understand the inner relationship between the geographical properties city areas and the relative mobility behaviour to generate synthetic mobility data. We consider a Generative AI approach and deep learning methods could suit perfectly for this task.

In particular, considering the natural networked structure of the cities, we aim to investigate Graph Neural Network approaches that might be able to include, in the learning phase, not only the geographical properties of an area but also its topological characteristics.

Moreover, the generation of synthetic mobility data can offer numerous benefits in terms of privacy, confidentiality, and proprietary concerns. However, it is crucial to balance two key factors: the utility of the generated data for analytical purposes and the need for the data to be sufficiently different from the original to maintain its integrity and avoid privacy issues. This could also unlock big companies’ release of more datasets if the datasets satisfy the privacy-by-design requirements.

This research proposal aims to explore the political economy of tourism in urban environments, with a specific focus on Madrid. The study intends to capture the sentiments and opinions of Madrid residents regarding the growth of tourism and its socio-economic impacts and voting behaviour. The primary method of data collection will be a comprehensive online survey, which seeks to understand the diverse views of the city’s population towards tourism-related changes.

A key aspect of this research is its approach to discerning the causal effects of tourism on public opinion. By leveraging the unexpected influx of tourists due to the rise of Airbnb, the study aims to isolate the direct influence of this increase in tourism on the attitudes and perceptions of the local residents.

This approach is particularly innovative as it uses the sudden change brought about by Airbnb as a natural experiment to understand the broader impacts of tourism. Overall, this study seeks not only to contribute to academic discussions on the political economy of tourism but also to provide practical insights for effective and sustainable tourism management in urban settings. By considering the preferences and opinions of Madrid’s residents, the research aims to support the development of tourism policies that are beneficial for both the city and its inhabitants.