This app was created as a prototype for our DHBW-project "concepts and operational scenarios of graph databases" where we analyzed and evaluated different graph databases.
Searching for an interesting use case with a very huge amount of publically available data, we discovered the Wikipedia-Game: We downloaded the wikipedia-dumps and parsed them using the graphipedia-script to import the Wikipedia-Pages (Nodes) and links between them (Edges) in our test winning graph databases, Neo4j and ArangoDB. For ArangoDB we wrote a multi-core python parser, which directly imports the data using arangoimp.
Using our app, you can calculate the shortest path between to Wikipedia-pages, demonstrating the power of graph databases.
Have fun!
- @pascalherrmann: Front-End (AngularJS), Node.js, Neo4j, Docker-Images
- @henp95: Back-End (PHP with Lumen from Laravel), ArangoDB, Data import with Python
We used Docker to simplifiy the installation process. All you have to do in order to build and start the required Docker images is the following:
docker-compose up
- Go To http://localhost:8081
docker-compose -f docker-compose-node.yaml up
- Go to http://localhost:8080
You might wanna apply some changes to docker-compose.yaml in order to change port allocation or language of the installed wikipedia.
