Graph database design principles, general principles and detail problem

Question

There are normal forms in the structure of a relational database to guide the design process. Are there similar principles that apply to the design of graphical databases such as neo4j?

In particular, I am puzzled by the granularity problem: I could create a graph database where most of the attributes are stored at the vertices (the contact has the attribute name, date of birth, ...), or I could store most of the data in the relationship (linking the contact with the relation " firtst name "with another vertex containing the actual information), or I could do both (then, of course, there are problems of consistency, but this can speed up the search time).

These problems are similar to my starting questions when I started developing SQL, and many of the problems could only be solved by gaining experience. You can still recommend a good practical book on the very basics of this topic or tell me where to find the general principles?

Answer 1

There is nothing that I saw that is perceived as truth, as normalization in relational databases. However, I would defend two things:

1) Model objects (nodes) as usual. Attributes of these objects are things that belong specifically to these entities and exist outside of relationships (compounds).

2) (), . , , () ().