The Louvian algorithm's usage is similar to the other algorithms. To stream the results, use the gds.louvain.stream procedure with a named projected graph as a parameter:
CALL gds.louvain.stream(<projectedGraphName>)
We can also save the results in the graph with the equivalent write procedure:
CALL gds.louvain.write(<projectedGraphName>, {writeProperty: <newNodeProperty>})
Using the Louvain algorithm on our graph leads the two usual partitions, with nodes A, B, and C on one side and nodes D to G on the other side. To see the differences between these algorithms, we will have to move to slightly larger graphs. We will see an example of this with Zachary's karate club in a later section.
