This project is for notebooks initiated, they migrate open data sources to the graph database Neo4J. More informations about open data can be find under the following link: (https://okfn.org).
The code in this project is developed with Python v3 and Jupyter v5.0.0.
In this moment exists only one notebook to bring open data to Neo4J. In the notebook DB_OpenData/Carsharing_Data.ipynb is demonstrated, how some relational csv data from OpenData-platform of Deutsche Bahn (DB) can be migrated to a Neo4J database.
To successfully execute the transformation on this notebook and migrate the data to a Neo4J database, you need to take the following steps:
- Clone this project
- Download the following csv-sources from the website.
- OPENDATA_BOOKING_CARSHARING.csv
- OPENDATA_CATEGORY_CARSHARING.csv
- OPENDATA_RENTAL_ZONE_CARSHARING.csv
- OPENDATA_VEHICLE_CARSHARING.csv
- Save the csv-sources in the subfolder "DB_OpenData/datasets"
- Create the subfolder "DB_OpenData/output"
- Execute the transformation in notebook DB_OpenData/Carsharing_Data.ipynb
- Modify the script DB_OpenData/scripts/loadDataToNeo4J.sh depend on the installation folder in your environment.
- Execute the script DB_OpenData/scripts/loadDataToNeo4J.sh. The script needs as parameter the path of the folder, in that the input files have been saved.
After you've migrated the data to Neo4J, you can also find an example to explore this data with python-based components in the notebook DB_OpenData/Explore_With_py2neo_NetworkX.ipynb
In the folder extensions you'll find custom extensions for Neo4J. Currently exists one extension (de.m.k.neo4j.ext), that prepares and provides data to visualize booking-informations as sankey-diagram. The component, that provides the data (JSON) for a sankey diagram can also be used for other types of nodes (see therefore the class SankeyDataResource).
I hope you've fun with the project. Feel free to contribute and give suggestions.