Using GDS to assess PageRank centrality in Neo4j

As usual, when using GDS, we need to define the projected graph that will be used to run the algorithm. In this example, we will run the PageRank algorithm on the directed graph, which is the default behavior of projected graphs. If you didn't do so in the previous section, you can use the following query to create the named projected graph out of the nodes with a Node label and relationships with the LINKED_TO type:

CALL gds.graph.create("projected_graph", "Node", "LINKED_TO")

We can then run the PageRank algorithm on this projected graph. The signature of the PageRank procedure in Neo4j is as follows:

CALL gds.pageRank(<projected_graph_name>, <configuration>)

The configuration map accepts the following parameters:

• dampingFactor: A float between 0 and 1 corresponding to the damping factor (d, in our Python implementation). The default value is 0.85.
• maxIterations: The...