Skip to content

This repo contains the data and queries from the Neo4j Connections:GDS demonstration of using the graph data science library with a retail dataset.

Notifications You must be signed in to change notification settings

jbaktir/GDS_Retail_Demo

 
 

Repository files navigation

Graph Data Science Retail Demo

April 28, 2020

This repo contains the data, load scripts, and queries that were demonstrated during the Connections: Graph Data Science presentation “What is Graph Data Science, and Neo4j’s GDS Library” presentation.

Database setup: To replicate the demo, create an empty Neo4j 4.0 database and install the 1.2 release of the Graph Data Science library. Copy the .csv files from the /data folder into your plugins directory, and execute the queries in load_data.cypher.
Or to install directly from this repository execute the queries in load_data_online.cypher

Running the algorithms: Each of the other folders (Customer Segmentation, Item Similarity, Item Centrality, and Exploratory Queries) contains the procedure calls and queries demonstrated during the presentation.

Bloom: The Bloom perspective presented is available in GDS_Retail.json -- to use it, import the perspective into Bloom 1.3.0 when connected to the Retail database.

Note: You may need to change the Community IDs specified in the Item Similarity, Ite, Centrality, and Exploratory Queries to match the community IDs in your database. They are assigned based on seeds pulled from your internal node IDs, so they may be different in your database.

About

This repo contains the data and queries from the Neo4j Connections:GDS demonstration of using the graph data science library with a retail dataset.

Resources

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published