Individual Mobility Analytics with Open Data and Software

Individual mobility is a crucial indicator of personal health and behavior. To assess individual mobility and activity, mobile sensing using global positioning systems (GPS) and surveys on smartphones are increasingly being employed, providing high spatial and temporal resolution data. This advancement has facilitated in-depth studies into the interactions between individuals' real-life behaviors, health, personal characteristics, and environments at a micro spatiotemporal scale. For such studies related to personal health and mobility, several open-source software packages (e.g., R/Python libraries) and open trajectory data (e.g., GPS data) have been developed.

In this presentation, we will introduce our R package development project for individual mobility analytics, focusing on GPS-based trajectory data processing and analysis at an individual level. The open-source package includes: (1) GPS data preprocessing, (2) construction of semantically enriched trajectory data from GPS data using automated methods for home detection, stop-move detection, and transport mode detection, and (3) computation of mobility indicators at two different aggregation levels (daily vs. by-person). We will also showcase case studies using open GPS data (such as Geolife) and datasets from other research projects. Furthermore, we will discuss the challenges faced during this project and outline its future directions.