Graph Measures & Metrics
The Wolfram Language supports a broad range of measures that characterize graphs, from simple measures, such as the number of vertices and edges that tell the size and sparsity of a graph, to vertex degrees, which tell how locally well-connected each vertex is. Other measures include the geodesic distances in a graph or centrality measures that give a measure of how central in the overall graph each vertex is; for example, PageRank and HITS are measures used to order web page importance as returned from a search engine.
Basic Measures
VertexCount — the number of vertices
EdgeCount — the number of edges
Degree Measures
VertexDegree — the number of edges for each vertex
VertexInDegree — the number of in-edges for each vertex
VertexOutDegree — the number of out-edges for each vertex
Distance Measures
GraphDistance — the length of the shortest path between two vertices
GraphDistanceMatrix ▪ VertexEccentricity ▪ GraphRadius ▪ GraphDiameter
Connectivity Measures
VertexConnectivity — the number of vertex-independent paths between two vertices
EdgeConnectivity — the number of edge-independent paths between two vertices
Centrality Measures
ClosenessCentrality — inverse average distance to every other vertex
BetweennessCentrality — fraction of shortest paths that pass through the vertex
DegreeCentrality ▪ EigenvectorCentrality ▪ KatzCentrality ▪ PageRankCentrality ▪ HITSCentrality ▪ RadialityCentrality ▪ StatusCentrality ▪ EdgeBetweennessCentrality
Reciprocity and Transitivity
GraphReciprocity — fraction of directed edges that are reciprocated
GlobalClusteringCoefficient — fraction of length-two paths that are closed
MeanClusteringCoefficient ▪ LocalClusteringCoefficient
Homophily, Assortative Mixing, and Similarity
GraphAssortativity — within-group connectivity minus between-group connectivity
VertexCorrelationSimilarity — correlation similarity between actors
MeanNeighborDegree ▪ MeanDegreeConnectivity ▪ VertexDiceSimilarity ▪ VertexJaccardSimilarity ▪ VertexCosineSimilarity