Simple Netflix-style Recommender System leveraging on Neo4j 3.1 and Spring, made for fun and educational purposes.

The sample movies are taken from the The Open Movie Database, http://www.omdbapi.com/. You can find the dataset in the file movies.json within the resource directory, or querying the REST application endpoint http://localhost:8080/findMovies.

There is also a csv file with the ratings of the users that watched those movies, ratings.csv You can obtain it as well querying the REST application endpoint http://localhost:8080/userRatings

Open the file recommendation.json. You will find an array of couples: the name/description of the recommendation technique and the corresponding Cypher query to run it against the graph db. An example is

{
  "name": "BEST ITALIAN COMEDIES",
  "query": "MATCH (c:Country)<-[:PRODUCTED_IN]-(n:Movie)-[:HAS_GENRE]->(g:Genre) where g.name = 'Comedy' and c.name = 'Italy' RETURN n.poster as poster LIMIT 25"
}

Note that the application expects the Cypher query to return the variable named poster. It is the image url and the application will use it to render the movie on the screen.

The challenge

1. Download Whitehall-flix application
2. Modify the application.properties and set the neo4j properties according to the Neo4j Server running your dataset. You should use the bolt protocol

For example

neo4j.uri=bolt://localhost:7687  
neo4j.username=neo4j  
neo4j.password=neo4j  

3.Modify the recommendation.json in order to provide recommendations. The challenge requires you to provide the following recommendations:

• MOST POPULAR movies in the dataset
• SIMILAR TO WHAT YOU WATCHED: the movies that should be recommended to the user according to her interests and preferences
• YOUR FRIENDS LOVED THOSE: the movies that should be recommended according to the friends interests

You can add more entries! Let your creativity flow! :D 4. Run the application with your favourite IDE or build it with
mvn clean install -Dmaven.test.skip=true and then run it with java -jar target/whitehall-flix-0.0.1-SNAPSHOT.jar

5. Take a look to http://localhost:8080 and have fun :D