Journal Article
Workshop: Analysing massive open human mobility data in R using spanishoddata, duckdb and flowmaps
AGIT Conference, 1:1, 229–231 (2025)
Abstract
Large-scale open human mobility datasets offer unprecedented research opportunities but present significant challenges in data acquisition, processing, and interactive visualization. In a 75-minute session, we showcased the Spanish Open Mobility Big Data case study and employed the R packages spanishoddata, duckdb, flowmapper, and flowmapblue to guide participants through reproducible methods for acquiring, analyzing, and visualizing high-resolution origin–destination flows. Participant feedback highlighted the dataset’s scale, accessibility, and ethical considerations, and generated questions spanning data representativeness, domain coverage, and DuckDB internals. In the next iteration of the workshop, we would consider extending the format to include a 90-minute tutorial segment plus an additional 60–90 minutes for hands-on data exploration and discussion to deepen engagement. We envision a follow-up series that integrates a technical deep-dive on DuckDB best practices (indexing, concurrency control, partitioning) with applied case studies using the Spanish mobility dataset, equipping attendees with both domain-specific insights and broadly applicable database skills.
Keywords: Spain, computational social science, data processing, software, teaching