Node Influence Measures

Node influence measures are used to rank the nodes in a network graph based on how influential they are in the whole network. The most common node influence measures used are:

Node Degree

shows how many connections the node has. This measure shows the local influence of the node as it calculates the connections it has to its immediate neighbors. There is a distinction between in-degree (connections going out of the node in directed graphs) and out-degree (connections coming in). InfraNodus, however, uses an undirected network by default, so it calculates a sum for the in- and out-degrees. 

To provide an analogy with the social networks, a node with a high degree is the one that "knows" many people.

Betweenness Centrality

betweenness centrality indicates how often the node appears on the shortest path between any two randomly chosen nodes in the graph. Measures the global influence of the node, because this measure will be higher for the nodes that connect different clusters of nodes together. It will be especially high for those nodes that serve as the bridges between two distinct communities (especially if there are no other nodes connecting them).

To use an analogy from social networks, a node with a high betweenness centrality will know many people from different communities. This node may not have too many connections, but if it's connects different groups with a high number of nodes within them together, then it will have a higher influence measure than another node with the same number of connections that are more local in nature.

Based on these two measures, there are additional measures that are used in InfraNodus to calculate the influence for every node:

Node Conductivity

Betweenness divided by Degree: which nodes can reach the most nodes in the different parts of the network faster with the least connections. Nodes with a high conductivity will tend to be better connected to the different distinct groups of nodes. They are reaching for global connections instead of the local ones. 

Node Locality

Degree squared divided by Betweenness: Local influencers with the least global connections. Nodes with high locality measure will tend to have more connections within their local group of nodes.

Node Diversivity 

Betweenness divided by Frequency: highest global connectivity with the least mentions - indicates the turning points that produce the narrative plot shifts.