In this project we test the use of Neo4j, a graph database, using the Movielens dataset. With Neo4j Desktop we perfom different query that you can find in the followinf folder. Two approaches have been proposed to analyze the data. The first approach is to run queries directly from the Neo4j Desktop. In the second approach we have developed a Spring Boot Application that accesses the database and returns the desired information. Project developed during the Database Design with New Generation Models course @ Department of Computer Science @ University of Bari "Aldo Moro" under the supervision of dr. Nicola Fanizzi.
By default, requirement for compilation are:
- JDK 11+
- Maven
Requirement for running application is:
- Neo4j Desktop 4+
You can download the dataset from here. We use the largest dataset with more than 20 milion ratings. Because of the size of the dataset we use the bulk import to load the data. More details here. Now we can use some example query you can find in the folder.
To create a jar file with dependencies including resource files:
$ mvn install
Application can be run from root folder by typing into terminal:
$ mvn spring-boot:run
Now from your browser you can typing (the params are simple example of data stored in Neo4j):
localhost:8080/titlesByNameContains/Toy Story (1995)
localhost:8080/allMovies
localhost:8080/allGenres
localhost:8080/allUser
localhost:8080/findUserById/3
This software is released under the MIT License. See the LICENSE file for details.