The four fundamental data constructs

As you may already know, the graph theory gives us many different graphs to work with. Graphs come in many different shapes and sizes, and therefore, Neo4j needed to choose a very specific type of data structure that is flexible enough to support the versatility required by real-world datasets. This is why the underlying data model of Neo4j, the labeled property graph, is one of the most generic and versatile of all graph models.

This graph data model gives us four different fundamental building blocks to structure and store our data. Let's go through them:

• Nodes: These are typically used to store entity information. In the preceding example, these are individual books, readers, and authors that are present in the library data model.
• Relationships: These are used to connect nodes to one another explicitly and therefore provide a means of structuring your entities. They are the equivalent of an explicitly stored and precalculated join-like operation in a relational database management system. As we have seen in the previous chapters, joins are no longer a query-time operation--they are as simple as the traversal of a relationship connecting two nodes. Relationships always have a type, a start and an end node, and a direction. They can be self-referencing/looping but can never be dangling (missing a start or end node).
• Properties: Both nodes and relationships are containers for properties, which are effectively name/value pairs. In the case of the nodes, this is very intuitive. Just like a record in the relational database world has one or more fields or attributes, so can the node have one or more properties. Less intuitive is the fact that relationships can have properties too (like additional columns in a join table). Properties are used to further qualify the strength or quality of a relationship and can be used during queries/traversals to evaluate the patterns that we are looking for.
• Labels: This was a fundamental data model construct that was added to Neo4j with version 2.0 at the end of 2013. Labels are a means to quickly and efficiently create subgraphs (categorize nodes). By assigning labels to nodes, Neo4j makes the data model of most users a lot simpler. There is no longer a need to work with a type property on the nodes or a need to connect nodes to definition nodes that provide meta-information about the graph. Neo4j now does this out of the box--and this is a huge asset, now and for the future. At the time of writing this book, labels are primarily used for the indexing and some limited schema constraints.
  However, in future, it is likely that the structural understanding that labels provide about the data stored in the graph will be used for other purposes such as additional schema, security, graph sharding/distribution, and perhaps others.
  In a relational database, a node with one label would be equivalent to a row (node) within a table.
  Neo4j allows you to set several labels per node, which has no semantic equivalence in the relational paradigm.
  Relations have only one label.

With these four data constructs, we can now start working with Neo4j.