Optimal way to design for neo4j from the start

Hi, if the target is to create a recommendation engine based on real-time product purchases and the data collection could be optimized for neo4j from the start.

  1. Are there any datasets I could look at to see how to properly collect the data for this purpose?
  2. What would the optimal way add the data to neo4j?

So it really depends on what you want the graph data model to look like. For instance, if you want to run queries focused on product pricing or order dates, then your data model would look completely different than if you wanted to query data about logistics and shipping of the orders.

One good place to start for something similar would be with the Northwind data set. We have a variety of topics for data modeling with this data set, and it's on product purchases.