Emergent Mind

Abstract

In this paper, we present a comprehensive survey of human-mobility modeling based on 1680 articles published between 1999 and 2019, which can serve as a roadmap for research and practice in this area. Mobility modeling research has accelerated the advancement of several fields of studies such as urban planning, epidemic modeling, traffic engineering and contributed to the development of location-based services. However, while the application of mobility models in different domains has increased, the credibility of the research results has decreased. We highlight two significant shortfalls commonly observed in our reviewed studies: (1) data-agnostic model selection resulting in a poor tradeoff between accuracy vs. complexity, and (2) failure to identify the source of empirical gains, due to adoption of inaccurate validation methodologies. We also observe troubling trends with respect to application of Markov model variants for modeling mobility, despite the questionable association of Markov processes and human-mobility dynamics. To this end, we propose a data-driven mobility-modeling framework that quantifies the characteristics of a dataset based on four mobility meta-attributes, in order to select the most appropriate prediction algorithm. Experimental evaluations on three real-world mobility datasets based on a rigorous validation methodology demonstrate our frameworks ability to correctly analyze the model accuracy vs. complexity tradeoff. We offer these results to the community along with the tools and the literature meta-data in order to improve the reliability and credibility of human mobility modeling research.

We're not able to analyze this paper right now due to high demand.

Please check back later (sorry!).

Generate a summary of this paper on our Pro plan:

We ran into a problem analyzing this paper.

Newsletter

Get summaries of trending comp sci papers delivered straight to your inbox:

Unsubscribe anytime.