Writing results to the graph

To visualize the results in the browser with the code we wrote in the previous section, we need to store the results in the graph, which can be achieved with the gds.labelPropagation.write procedure:

CALL gds.labelPropagation.write(
    "projected_graph_with_properties", {
        seedProperty: "knownCommunity",
        writeProperty: "lp"
    }
)

Algorithms such as Label Propagation are a good example of how graph algorithms have already been used in more classical machine learning models. Indeed, Label Propagation is used both for classification and regression in machine learning, where the propagation is performed through a similarity matrix (instead of the adjacency matrix discussed in the previous chapter).

For now, we will focus on another important algorithm for community detection: the Louvain algorithm.