Human mobility behavior in COVID-19: A systematic literature review and bibliometric analysis

Abstract

This article maps the scientific literature in human mobility behavior in the context of the current pandemic. Through bibliometrics, we analyze the content of published scientific studies indexed on the Web of Science and Scopus during 2020. This enables us the detection of current hotspots and future directions of research. After a co-occurrence of keywords and evidence map analysis, four themes are identified, namely, Land Transport - Operations, Land Transport - Traffic Demand, Air Transport and Environment. We show how air transportation- and environmental-related studies tend to be more mature research whereas the understanding of changes in travel behavior (e.g., telecommuting, preventive measures or health protection behavior) tends to be immature. By using a topic modeling approach, we identify multiple sub-themes within each theme. Our framework adopts a smart literature review approach that can be constantly updated, enabling an analysis of many articles, with little investment of the researcher's time, but also provides high degree of transparency and replicability. We also put forth a research agenda that can help inform and shape transport policy and practice responses to COVID-19.

Keywords: Bibliometrics; COVID-19; Text mining; Travel behavior.

Fig. 1

Flowchart of the experimental procedure to identify the emerging COVID-19 research trends in human mobility behavior..

Fig. 2

Publication venues.

Fig. 3

Publication venues and monthly number of articles.

Fig. 4

Keyword co-occurrence network of COVID-19 literature in the human mobility behavior area. The network graph depicts the top 50 keywords. The size of the node indicates frequency and edges represent co-occurrence relationships. Colors are used to highlight node clusters using Louvain method for community detection (Blondel et al., 2008). (For interpretation of the references to color in this figure legend, the reader is referred to the web version of this article.)

Fig. 5

Evidence maps.

Fig. 6

Number of sub-themes vs. perplexity score (lower is better) and coherence (higher is better).