Constructing a Recommendation Graph with H&M Personalization Dataset
While Neo4j is great for building knowledge graphs, it would be prudent to look at how we model the data. A good data persistence model can make data retrieval optimal and handle large loads better. In this chapter, we will take a step back to look at what constitutes a knowledge graph and how a different look at data modeling with a Neo4j data persistence approach can help build more powerful knowledge graphs. You might need to revisit the approaches defined in Chapter 3, which will enable you to build a knowledge graph with Personalized Fashion Recommendations (H&M personalization) data.
We will cover these topics in this chapter as we tackle data modeling evolution:
- Modeling a recommendation graph with the H&M Personalization dataset
- Optimizing for recommendations: Best practices in graph modeling
