Find the supplementary repo here or watch the videos!
A location-based social network dashboard for privacy-aware analysis based on LBSN structure, a Docker-based Postgres HyperLogLog implementation from Dunkel, Löchner, Krumpe et al. for LBSN analysis. More info here.
Disclaimer: This is a prototype for research purposes and not thought for production (subject to SQL-injection)!
- Backend consisting of Python web framework (fastapi) excluding docker containers from LBSN structure
- Frontend based on Leaflet and Geoman ready-to-deploy with plugin options (e.g. GeoJSON)
Use my Fast-Instagram-Scraper to retrieve Data e.g. from Instagram. It's fast, easy to use and quickly read into the privacy-aware DB.
- Set up LBSN Docker container with pgadmin (good for quickly checking SQL statements but not necessary) and HLL-DB
- Download some data from any LBSN such as Instagram. If you use Fast-Instagram-Scraper you can use lbsntransform to read the data into the DB with the following command, automatically using
instagram-mapping-for-fast-instagram-scraper.py(thanks and credits to Alexander Dunkel!)
lbsntransform --origin 13 --input_path_url "path/to/data/fast-instagram-scraper/your-area-of-interest" --file_input --dbpassword_output "eX4mP13p455w0Rd" --dbuser_output "postgres" --dbserveraddress_output "127.0.0.1:25432 " --dbname_output "hlldb" --dbformat_output "hll" --dbpassword_hllworker "eX4mP13p455w0Rd" --dbuser_hllworker "postgres" --dbserveraddress_hllworker "127.0.0.1:25432 " --dbname_hllworker "hlldb" --include_lbsn_objects "origin,post" --file_type "json" --mappings_path "/mappings/" --include_lbsn_bases hashtag,place,date,community,latlng
- Clone repo
- Install Python dependencies
- Start Docker container with HLL-DB
- Adjust DB connection details in
main.py, remove my bounding boxes for Bonn - Start backend with
python app.py - Go to localhost:8000