profile_urban_region
Use this skill when the user wants to identify urban functional regions such as residential, commercial, transit, leisure, or mixed-use areas by combining trajectory activity and POI evidence.
Files6
View on GitHubSKILL.mdskills/custom/spatiotemporal_trajectory/profile_urban_region/SKILL.md
GitHubprofile_urban_region
Use this Skill to classify region functions from activity time signatures and optional POI category distributions.
Command
cd /mnt/skills/custom/profile_urban_region
python3 scripts/profile_urban_region.py \
--input /path/to/cleaned_points.jsonl \
--output-dir /path/to/output \
--time-col utc \
--lat-col Latitude \
--lon-col Longitude \
--user-col user_id \
--poi-input /path/to/pois.csv \
--poi-category-col "Venue Category Name"
Outputs
region_profiles.jsonlsummary.json
Algorithm
- Geohash region aggregation
- Morning/workday/evening/night activity signature
- POI category keyword scoring
- Functional label ranking with mixed-use fallback
Next Skills
Use profile outputs with $fuse_spatial_evidence, or combine with hotspots and anomaly outputs for region portraits.