A survey on Human Mobility and its applications

TL;DR

This survey reviews human mobility research, covering trajectory analysis, network modeling, and applications like congestion prediction, highlighting challenges with new cellular data and proposing new approaches.

Contribution

It provides a comprehensive overview of human mobility studies, comparing data types and addressing challenges with new cellular mobility data.

Findings

Trajectory measures reveal movement patterns and predictability.

Graph models help analyze network flow and influence.

Cellular data introduces new challenges in data handling.

Abstract

Human Mobility has attracted attentions from different fields of studies such as epidemic modeling, traffic engineering, traffic prediction and urban planning. In this survey we review major characteristics of human mobility studies including from trajectory-based studies to studies using graph and network theory. In trajectory-based studies statistical measures such as jump length distribution and radius of gyration are analyzed in order to investigate how people move in their daily life, and if it is possible to model this individual movements and make prediction based on them. Using graph in mobility studies, helps to investigate the dynamic behavior of the system, such as diffusion and flow in the network and makes it easier to estimate how much one part of the network influences another by using metrics like centrality measures. We aim to study population flow in transportation networks using mobility data to derive models and patterns, and to develop new applications in predicting phenomena such as congestion. Human Mobility studies with the new generation of mobility data provided by cellular phone networks, arise new challenges such as data storing, data representation, data analysis and computation complexity. A comparative review of different data types used in current tools and applications of Human Mobility studies leads us to new approaches for dealing with mentioned challenges.