Graph Global Metrics

By for November 11, 2016

38 views
19 downloads

Language

Report Abuse
Graph Global Metrics is used to compute several summary statistics for a given graph.
This is a part of a series of custom modules based on the CRAN igraph package. Graph Global Metrics is used to compute several summary statistics for a given graph. Currently, computed global metrics are: number of nodes, number of edges, maximum degree, minimum degree, average degree, diameter, average path length, edge density, largest clique size, number of largest cliques, number of maximal cliques, number of triangles, cluster coefficient, number of connected components, largest component size. For directed graphs, the number of connected components and largest component size are calculated for both weakly and strongly connected components. Here is an example showing how to use this custom component. ![enter image description here][1] In this example, network.csv was generated using available code as described in “Lancichinetti, Andrea, Santo Fortunato, and Filippo Radicchi: Benchmark graphs for testing community detection algorithms. Physical Review E 78.4 (2008): 046110”. The input dataset needs to be represented in two columns as a list of edges where each value in each column represents a Node Id and each row represents an Edge, as shown here. ![enter image description here][2] This is what the output of the module looks like. Here the output was generated in column format. ![enter image description here][3] [1]: https://alvilcek.blob.core.windows.net/azuremlcustommodules/Azure%20ML%20Custom%20Modules/documentation/Graph_Global_Metrics_1.PNG [2]: https://alvilcek.blob.core.windows.net/azuremlcustommodules/Azure%20ML%20Custom%20Modules/documentation/Graph_Global_Metrics_2.PNG [3]: https://alvilcek.blob.core.windows.net/azuremlcustommodules/Azure%20ML%20Custom%20Modules/documentation/Graph_Global_Metrics_3.PNG