Human Mobility

Human mobility is a fascinating phenomenon. While individual trajectories are difficult to predict (to some extent), human flows exhibit more regular patterns typical of complex systems.
Recently, it has been shown that human mobility is more predictable than expected, triggering the interest of a wide scientific community.

Human mobility prediction Learn more »


We have shown that it is possible to predict the next locations where an individual will move to, just by looking at his/her historical time series of movements and the one of his/her social contacts.
Our work has been awarded the first prize at the Nokia Mobile Data Challenge in 2012.



References:

Exploiting Big Data for personalized routing in smart cities Learn more »


We have developed a routing system that leverages multiple data sources coming from the city, to provide personalized routing based both on a balance between personal preferences and common interest.
Our work has been finalist at the Telecom Big Data Challenge in 2014, with an application to city of Milan.




References:

Identifying integration and segregation of human flows within cities Learn more »


Between April and July 2019 we participated and won, together with other research teams, to the Future Cities Challenge organized by Foursquare and an international team of scientists. Big data about human mobility was provided for 10 megacities around the world.

Increasing evidence supports the fact that cities are complex systems, with a broad spectrum of structural and dynamical features which lead to unexpected and emerging phenomena. Understanding urban dynamics at individual level -- but also as the outcome of collective human behaviour -- will open the doors to uncountable applications ranging from enhancing the sustainability and the resilience of the city to improving health and well-being of its inhabitants.

Here, we use a unique data set of longitudinal human flows provided by Foursquare, a leader platform for location intelligence, to characterize the functional organization of a city. First, we build multidimensional network models of human flows corresponding to different types of activities across time. We quantify the efficiency of flow exchange between areas of a city in terms of integration and segregation, respectively. Results reveal unexpected complex spatio-temporal patterns that allow us to gain new insight on the function of 10 megacities worldwide. We discover that large cities tend to be more segregated and less integrated, and that human flows at different hours of the day or between different types of activities enable the identification of different ``cities within the city'' which indeed show clear dissimilarities in terms of both functional integration and segregation.

Our analysis provides new insights on how human behaviour influences, and is influenced by, the urban environment and, as an interesting byproduct, to characterize functional (dis)similarities of different metropolitan areas, countries, and cultures.


Figure: Modeling Structure and Function of Urban Systems. Left: Urban structural backbone of the 10 megacities considered here, as described from their street networks (data obtained from Open Street Map. Middle: Urban functional networks described by the Foursquare data. The nodes are obtained by dividing the area analysed into cells of 500m x 500m. The edges are subsequent check-ins that might be between activities of the same type (intra-links: e.g. Food-Food, Tourism-Tourism) or different types (inter-links: e.g. Food-Tourism, Food-Sport). The collection of layers and inter-layer flows defines a multilayer network, i.e., a multidimensional functional representation of the urban areas. Right: The mobility flows between areas are captured as the edges' weights. In the example, describing New York City, we can observe the different spatial distribution of flows between and across different activity layers.

Figure: Disentangling human flows. We illustrate here, for four of the ten cities studied in this paper, the strikingly distinct views on the functional organization of a city extracted by isolating intra- or inter-layer flows. These maps outline the different cities within the city which we disentangle by decoupling the urban flows into activity-aware multilayer networks.


References:

Human mobility models Learn more »


We have proposed a mobility model at country level and we have intertwined it with a model for the diffusion of awareness about the spreading of a disease. We have shown that standard countermeasures, such as quarantine, might be less effective in containing the disease spreading than targeted information campaigns.
Our work has been awarded the first prize at the Orange D4D Challenge in 2012, with an application to developing countries.


Countermeasures similar to the one proposed in our work have been successfully applied to boost Ebola awareness in some countries of West Africa, to dramatically reduce the spreading of this deadly disease. [link1] [link2]


In a later study, we have developed a higher-order Markov mobility model, that we have named adaptive memory. This model allows to account for the waiting time of individuals conditional to their historical movements while dramatically reducing the number of spurious mobility patterns affecting more traditional Markovian models.




References: