This software allows to generate, visualize, and analyze attributed networks having a community structure. It is provided as is, for demonstration only. This URL should remains private as long as the article has not been accepted.

Communities

• K
• Number of communities
• n
• Number of vertices
• Nb. Rep.
• Number of representents in each communities. The higher is the value, the slower is the computation
• Theta
• Percentage of vertices assigned to a random community. The higher is this value, the less likely the community will be homogeneous w.r.t. the attributes

Attributes

• Nb. Attr.
• Number of attributes associated to the vertices
• Dev. i
• Standard deviation of the ith attribute

Edges

• Edges Within
• Maximum number of within community edges added to a newly inserted vertex
• Edges Between
• Maximum number of between community edges added to a newly inserted vertex
• MTE
• Minimum number of edges in the resulting graph (up to a graph where communities are cliques)

Attribute measures

• Observed homophily
• Ratio of edges connecting similar vertices w.r.t. their attribute values
• Expected homophily
• Ratio of pair of similar vertices among all possible pairs of vertices
• Within inertia
• Measure of the dispersion of the attribute values inside the communities. A low within inertia indicates that the communities are highly homogeneous w.r.t. the attribute values

Structural measures

• Modularity
• The partition modularity measure has defined by M.E.J. Newman
• Average clustering coefficient
• Average clustering coefficient in the graph
• Random clustering coefficient
• Clustering coefficient in a Erdös-Renyi random graph having the same number of vertices and edges
• Average degree
• Average number of neighbours of the vertices
• Average shortest path length
• The average minimum number of hops required to reach two arbitrary vertices. It is not computed when the graph is formed by several disconnected components
• Diameter
• Length of the longest shortest path between any pair of vertices
• Nb. edges between
• Number of edges connecting two vertices belonging to different communities
• Nb. edges within
• Number of edges connecting two vertices belonging to the same community
• Nb. edges
• Total number of edges in the graph

• Output
• To output the generated graph, press the button next to the random seed and select a file name. The file extension is .graph and is splitted in three sections: the parameters, the vertices and the edges.
• Parameters
• This section starts with the line # Parameters. The parameters are output with a line starting with # followed by the parameter name and its value.
• Vertices
• This section starts with the line # Vertices. Each consecutive lines describes a vertex. A line consists of an integer corresponding to the vertex id, the list of its attribute values separated by a ; and an integer corresponding to the vertex community id. Each value is separated by a ;.
  Example The line 1;0.31;0.49;2 corresponds to the vertex with id 1 associated to two attributes having values 0.31 and 0.49 and being in community 2.

• Edges
• This section starts with the line # Edges. Each consecutive lines corresponds to an edge. A line consists of two vertex ids separated by a ;.
  Example The line 1;4 corresponds to an edge between the vertex with id 1 and the vertex with id 4.

• Christine Largeron
• University of Saint-Étienne, France
• Pierre-Nicolas Mougel
• University of Saint-Étienne, France
• Reihaneh Rabbany
• University of Alberta, Canada
• Osmar Zaïane
• University of Alberta, Canada

Download ANC_Generator jar file from her : Download

Download ANC_Generator SCALA source file from her : Download

This software is based on an algorithm described in article : Generating Attributed Networks with Communities by Christine Largeron, Pierre-Nicolas Mougel, Reihaneh Rabbany and Osmar R. Zaïane published in PLOS ONE.