What is a graph database?

ChatGPT describes it this way:

A graph database is a type of database that stores data in the form of nodes (also known as "entities" or "nodes") and relationships (also known as "edges"). Each node represents an object or entity, while each relationship represents the connection between two nodes. This type of data structure is particularly useful for applications built on network data, such as social networks, recommender systems, and complex queries. Graph databases allow relationships between data to be easily stored and analyzed, making them particularly useful for processing unstructured data.

Accordingly, the HIVE blockchain could be perfectly modeled in such a graph database!

That's exactly what we're trying to do right now. Our current system works with similarity searches based on posts that people have interacted with. To do this, we record on our frontend Hive Discover, among other things, who spends how long on which post and calculate the interest in it. As a result, we only have to calculate the similarity and the interest and can create a personalized feed.
Even though this already works relatively well and the interactions are asynchronously encrypted on our servers (Only when a feed is requested, the activities are briefly decrypted. More information in this post), we find that it does not yet fit into Web3, as it is not yet fully transparent.
The new idea now would be to model the HIVE blockchain in Neo4J's graph database and thus be able to propose content completely transparently based on follows, votes, comments, etc. to be able to suggest content. Then every user would know exactly what data is being used and the whole application would be executable in a decentralized way by different parties!

What's your opinion on this one? Would it be an improvement or just another redirection of the problem and are you at all interested in such a change? - Feel free to write it in the comments!
Action-chain