/rudik-docker

Rudik Dockerisation

Primary LanguageJava

Rudik Dockerisation

This guide is to help you set up and run the Rudik project (https://github.com/stefano-ortona/rudik) using Docker Compose. The setup involves:

ontotext: A GraphDB service.

rudik: The Rudik service that runs the rule mining application.

Explanation

ontotext: Runs GraphDB and exposes it on port 7200. After the build, you should create a new repository on GraphDB and upload your RDF graph. Afterwards, you should provide it's SPAQRL endpoint to the related field in rudik/Configuration.xml. For SPAQRL endpoint URI, you can inspect GRAPHDB SPAQRL query page after you run a query. You may find it under the post request.

rudik: The main Rudik service.

Important Notes

  • Rudik runs manually in Docker. To discover the rules you should run below code in the rudik image (please wait until ontotext is available on localhost:7200):
java -cp target/rule_miner-0.0.1-SNAPSHOT.jar asu.edu.rule_miner.rudik.App
  • Target relation and subject & object types are hard-coded in App.java. Please change them accordingly.

  • Negative and positive rule results can be found under "negative_horn_rules.txt" and "positive_horn_rules.txt" in the rudik image file section.

Prerequisites

Docker installed on your machine.

Docker Compose installed on your machine.

Running the Setup

1- Build the rudik image:

docker-compose build rudik

2- Start the services:

docker-compose up