Once we have defined the input for our graph algorithm procedure, the projected graph, we also have to decide what the plugin will do with the results. There are three possible options:
- Stream the results: The results will be made available as a stream, either in the Neo4j browser or readable from the Neo4j drivers in different languages (Java, Python, Go, .NET, and so on).
- Write the results to the Neo4j graph: If the number of affected nodes is really big, streaming the results might not be a good option. In that scenario, it is possible to write the results of the algorithm into the initial graph; for instance, adding a property to the affected nodes or adding a relationship.
- Mutate the projected graph: The algorithm results are saved as a new property in the in-memory graph. They will have to be copied to the Neo4j graph later on, using the gds.graph.writeNodeProperties procedure.
In this book, we will work with static projected graphs...
