The betweenness centrality is defined for each node in a network. It measures how likely an arbitrary shortest path in the network will go through the node. A node with a high betweenness centrality is likely to sit in the middle of two large communities, or sub-networks, hence the name betweenness.
In CiteSpace, a node with a high betweenness centrality is shown with a purple ring. The thickness of the purple ring depicts the value of the betweenness centrality.
The use of betweenness centrality in CiteSpace is guided by the structural hole theory. The theory was originally developed for social networks. An insightful observation is that the connectivity, or the lack of it, can guide us to find the most valuable nodes in a network. CiteSpace builds on top of these theories to detect boundary-spanning potentials and novel brokerage connections in scholarly publications.
By default, CiteSpace automatically computes the betweenness centrality of all the nodes in a network, provided the size of the network is less than 350. You can modify the default setting: Preferences > Defer the calculation of betweenness centrality.

References