A Python script that converts ConceptNets CSV data a CSV input that can be fast imported into Neo4J
Python
Switch branches/tags
Nothing to show
Clone or download
Fetching latest commit…
Cannot retrieve the latest commit at this time.
Permalink
Failed to load latest commit information.
.gitignore
LICENSE
README.md
conceptnet_to_neo4j.py

README.md

conceptnet_neo4j

A Python script that converts ConceptNets CSV data a CSV input that can be fast imported into Neo4J

Neo4J is a useful graph database for performing complex graph queries, and thus is quite suitable for a graph database such as ConceptNet.

As Neo4J has a very useful ./neo4j-import tool, converting ConceptNet into a suitable input for this can enable a user to import ConceptNet in less than a minute.

This script takes in the extracted concept_flat_csv folder, and exports a node.csv and relationship.csv file which can then be imported into Neo4J using ./neo4j-import.

After running, you can import it into conceptnet using something like:

./neo4j-import --into "folder you want to store db in" --nodes "nodes.csv" --relationships "relationships.csv"