This is a demo on how to catch some people trying to commit transit fraud

The London Underground network has been loaded as a graph into Spanner Graph with the following relationships

Oyster Transit Cards are linked to people, who have addresses and the rides are stored in the database. It is possible to find fraud for a card that has swiped at a location that it could not possibly reach given the minimum transit time.

Once an entry point to a fraud investigation it is possible to link that card to other fradulent cards and discovery interesting relationships.

gcloud auth application-default login
make instancecreate
make loadschema

python3 -m venv .venv
source .venv/bin/activate
pip install -r requirements.txt
./load_data.py

./app.py

Queries

You can also run a raw query to receive back a JSON object detailing the paths

curl -s -X POST  -d '(Station{id: 155})-[r:ROUTE]->{1,4}(t:Station)' http://localhost:5000/raw |jq