Graph databases have gained increasing attention in the last few years. Data models built from graphs bring together the simplicity of document-oriented databases and the clarity of SQL tables. Among others, Neo4j is a database that comes with a large ecosystem, including the database, but also tools to build web applications, such as the GRANDstack, and tools to use graph data in a machine learning pipeline, as well as the Graph Data Science Library. This book will discuss those tools, but let's first start from the beginning.
Talking about graph databases means talking about graphs. Even if you do not need to know all the details about graph theory, it’s always a good idea to learn some of the basic concepts underlying the tool you are using. In this chapter, we will start by defining graphs and giving some simple and less simple examples of graphs and their applications. We will then see how to move from the well-known SQL tables to graph data modeling. We’ll conclude by introducing Neo4j and its building blocks, and review some design principles to understand what can and can’t be done with Neo4j.
This chapter will cover the following topics:
- Graph definition and examples
- Moving from SQL to graph databases
- Neo4j: the nodes, relationships, and properties model
- Understanding graph properties
- Considerations for graph modeling in Neo4j
