This repository contains both Jupyter notebooks for solving a link prediction problem using Neo4j’s Graph Data Science Library and scikit-learn.
The associated developer guide for this repository can be found at neo4j.com/developer/graph-data-science/link-prediction/scikit-learn/
You can run the examples by following the instructions below:
git clone https://github.com/neo4j-examples/link-prediction.git
cd link-predictionYou can launch the notebooks and Neo4j by using the following command:
docker-compose upThis will spin up a Neo4j server and Jupyter session. The terminal output will look a bit like this:
...
link-prediction-jupyter | [I 05:11:37.148 NotebookApp] The Jupyter Notebook is running at:
link-prediction-jupyter | [I 05:11:37.148 NotebookApp] http://5bf4a4a75310:8888/?token=06ce9ad89428f765734540f3f283cc0d3ad5b1fe0c51746a
link-prediction-jupyter | [I 05:11:37.148 NotebookApp] or http://127.0.0.1:8888/?token=06ce9ad89428f765734540f3f283cc0d3ad5b1fe0c51746a
link-prediction-jupyter | [I 05:11:37.148 NotebookApp] Use Control-C to stop this server and shut down all kernels (twice to skip confirmation).
link-prediction-jupyter | To access the notebook, open this file in a browser:
link-prediction-jupyter | file:///home/jovyan/.local/share/jupyter/runtime/nbserver-6-open.html
link-prediction-jupyter | Or copy and paste one of these URLs:
link-prediction-jupyter | http://5bf4a4a75310:8888/?token=06ce9ad89428f765734540f3f283cc0d3ad5b1fe0c51746a
link-prediction-jupyter | or http://127.0.0.1:8888/?token=06ce9ad89428f765734540f3f283cc0d3ad5b1fe0c51746a
...
link-prediction-neo4j | Starting Neo4j.
link-prediction-neo4j | 2020-08-20 05:11:44.729+0000 INFO ======== Neo4j 4.0.6 ========
link-prediction-neo4j | 2020-08-20 05:11:44.734+0000 INFO Starting...
...
link-prediction-neo4j | 2020-08-20 05:11:56.135+0000 INFO Remote interface available at http://localhost:7474/-
The Jupyter notebook is available at http://localhost:8888. You’ll need to enter the token from the output the first time that you launch it.
-
Neo4j is available at http://localhost:7474/
If you want to view the notebooks online, you can find them in the following locations:
React
Typescript
Django
Laravel
Facebook
Microsoft
Google
Alibaba
D3
Tencent