luisehica / gds_retail_demo

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

Graph Data Science Retail Demo

November 17, 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 1.4 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.

GraphSAGE: I've updated this repo with a quick demonstration of how to use GraphSAGE to learn an embedding, predict embeddings for data in your graph, and use the KNN algorithm to build a similarity graph, along with a Bloom perspective.

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.