The movements of individuals within and among cities influence critical aspects of our society, such as wellbeing, the spreading of epidemics, and the quality of the environment. When information about mobility flows is not available for a particular region of interest, we must rely on mathematical models to generate them. In this I first introduce the Gravity Model, a parametric model that can be used to predict flows among cities. I will show that such model is equivalent to a one layer classifier, i.e. a Perceptron. I will then study and apply a more general and effective model, Deep Gravity, that can generate flow probabilities exploiting many features (e.g., land use, road network, transport, food, health facilities) extracted from voluntary geographic data, and uses deep neural networks to discover nonlinear relationships between those features and mobility flows. I will show that Deep Gravity achieves a significant increase in performance, especially in densely populated regions of interest, with respect to the classic gravity model and models that do not use deep neural networks or geographic data.
The movements of individuals within and among cities influence critical aspects of our society, such as wellbeing, the spreading of epidemics, and the quality of the environment. When information about mobility flows is not available for a particular region of interest, we must rely on mathematical models to generate them. In this I first introduce the Gravity Model, a parametric model that can be used to predict flows among cities. I will show that such model is equivalent to a one layer classifier, i.e. a Perceptron. I will then study and apply a more general and effective model, Deep Gravity, that can generate flow probabilities exploiting many features (e.g., land use, road network, transport, food, health facilities) extracted from voluntary geographic data, and uses deep neural networks to discover nonlinear relationships between those features and mobility flows. I will show that Deep Gravity achieves a significant increase in performance, especially in densely populated regions of interest, with respect to the classic gravity model and models that do not use deep neural networks or geographic data.
A Deep Gravity model for mobility flows generation
SARAN GATTORNO, GIANCARLO
2022/2023
Abstract
The movements of individuals within and among cities influence critical aspects of our society, such as wellbeing, the spreading of epidemics, and the quality of the environment. When information about mobility flows is not available for a particular region of interest, we must rely on mathematical models to generate them. In this I first introduce the Gravity Model, a parametric model that can be used to predict flows among cities. I will show that such model is equivalent to a one layer classifier, i.e. a Perceptron. I will then study and apply a more general and effective model, Deep Gravity, that can generate flow probabilities exploiting many features (e.g., land use, road network, transport, food, health facilities) extracted from voluntary geographic data, and uses deep neural networks to discover nonlinear relationships between those features and mobility flows. I will show that Deep Gravity achieves a significant increase in performance, especially in densely populated regions of interest, with respect to the classic gravity model and models that do not use deep neural networks or geographic data.File  Dimensione  Formato  

Saran_Gattorno_Giancarlo.pdf
accesso riservato
Dimensione
3.37 MB
Formato
Adobe PDF

3.37 MB  Adobe PDF 
The text of this website © Università degli studi di Padova. Full Text are published under a nonexclusive license. Metadata are under a CC0 License
https://hdl.handle.net/20.500.12608/45496