logo
ResearchBunny Logo
A Deep Gravity Model for Mobility Flows Generation

Computer Science

A Deep Gravity Model for Mobility Flows Generation

F. Simini, G. Barlacchi, et al.

Discover Deep Gravity, a groundbreaking model that utilizes deep neural networks and geographic data to accurately predict human mobility flows. This research, conducted by Filippo Simini, Gianni Barlacchi, Massimilano Luca, and Luca Pappalardo, outperforms traditional gravity models, particularly in urban areas, all while employing explainable AI techniques for better understanding of the predictions.

00:00
00:00
Playback language: English
Abstract
This work introduces Deep Gravity, a model leveraging deep neural networks and diverse geographic data to generate human mobility flow probabilities. Experiments in England, Italy, and New York State demonstrate Deep Gravity's superior performance, particularly in densely populated areas, compared to traditional gravity models. The model exhibits good generalization capabilities and uses explainable AI techniques to interpret predictions.
Publisher
Nature Communications
Published On
Nov 12, 2021
Authors
Filippo Simini, Gianni Barlacchi, Massimilano Luca, Luca Pappalardo
Tags
Deep Gravity
human mobility
neural networks
geographic data
explainable AI
gravity models
urban population
Listen, Learn & Level Up
Over 10,000 hours of research content in 25+ fields, available in 12+ languages.
No more digging through PDFs, just hit play and absorb the world's latest research in your language, on your time.
listen to research audio papers with researchbunny