Mining dockless bikeshare and dockless scootershare trip data
summary
City permit data plus company monthly reports and public APIs feed Python scripts that produce concrete answers not abstracts: a random forest classifier trained on Capital Bikeshare labels docks vs casual trips and predicts ~60% of dockless bikes and ~65% of scooters resemble members, cleaning removes ~33% of scooter records and ~29% of bike records, idle-time analysis shows scooters have much shorter idle durations, hotspot finding uses DBSCAN clustering (25 m epsilon, ward-tuned minPts) and shiny visualizations to map candidate rack sites.
AI-written summary · about
chapters (6)
- 0:00Dockless bikeshare and scootershare intro
- 6:24Capital Bikeshare vs dockless vehicles
- 12:12Idle time in public spaces
- 19:34User behavior patterns
- 25:11Parking hotspots
- 32:06Q&A
related talks
43mPython at City Scale
34mCityBikes: bike sharing networks around the world
42m2017 DevFestMN Sara Robinson
30mAnalyzing 33 million bike trips with BigQuery
33mAnalyzing 33 million bike trips with BigQuery
25mAtlas of Popular Transport: Democratizing mobility data
41mAnalysing and predicting inner-city parking space occupancy
28mRevenue maximisation from PEDL trips using network analysis and geospatial mapping