A Mechanistic Data-Driven Approach to Synthesize Human Mobility Considering the Spatial, Temporal, and Social Dimensions Together

Round 1

Reviewer 1 Report

This is an interesting paper that addresses a relevant problem, filling a clear gap in the literature.

I only have a couple of minor corrections to suggest and I'm otherwise happy to recommend the paper for publication.

Minor corrections:

• It may be useful to list all the parameters of your model in a table, perhaps highlighting which ones are preset (e.g., the social factor) and which ones are learned from the data.
• It would be also beneficial for the reader to quickly explain the rationale leading to some parameters being fixed (e.g., the social factor being 0.2) instead of just mentioning the reference.

Other comments:

whit respect to real-world trajectories -> with respect to real-world trajectories

three mechanisms affect the realism of the others -> three mechanisms affecting the realism of the others

Author Response

Please see the attachment.

Author Response File: Author Response.pdf

Reviewer 2 Report

Dear authors,

Please find my comments and suggestions in the attachment.

Comments for author File: Comments.pdf

Author Response

Please see the attachment.

Author Response File: Author Response.pdf

Reviewer 3 Report

This paper proposes a Spatial, Temporal, and Social Exploration and Preferential Return (STS-EPR) modeling framework which is a mechanistic, generative model to capture spatial, temporal, and social aspects of human mobility. It is a well-organized paper with an interesting topic that can be useful for studying processes of considering both sociality and mobility. Overall, this is a decent paper with some areas for further improvements.

1. Please give the full name of STS-EPR when it first appears in the abstract of this paper.
2. In Section 4.2, please explain why it sets ρ = 0.6, γ = 0.21, and α = 0.2. What would be the impacts on the analysis results if these parameters were set at different values? What are the guidelines of choosing these parameter values?
3. In Section 4.3, why did this study use a gravity model with a distance exponent of 2? Was it calibrated based on some real-world data?
4. In Section 5.1, Foursquare checkin data are unlikely to cover all stops in an individual’s trajectory. Is it a problem to use Foursquare checkin data to build individual trajectories?
5. The real trajectory and the synthetic trajectory in Figure 2 appear to be quite different. In this case, how can we trust synthetic trajectories generated by STS-EPR can represent real trajectories at the disaggregate level to an acceptable level? The analysis results reported in Table 2 on p. 14 appear to be for comparisons at an aggregate level.
6. Why does the paper reference Figure 7 on p. 11 before it references Figure 6 on p. 12? Should they be numbered in order?
7. There are two Table 2s in this paper: one on p. 13 and another on p. 14. It appears that Table 2 on p. 13 should be renumbered as Table 3.
8. In Section 6.4, this paper reports that “The choice of the spatial tessellation impacts the results. (line 418)” However, this paper falls short of providing guidelines of choosing an appropriate spatial tessellation for a study.  This makes it hard to assess the reliability of the proposed STS-EPR framework.
9. There are some typos in this paper. For example,

Line 8: Should “… whit respect to …” be “…with respect to …”?

Line 91: Should “… an agent return to …” be “… an agent returns to …”

Author Response

Please see the attachment.

Author Response File: Author Response.pdf