Social network analysis (SNA) investigates the connections and relationships between people or groups within a social network. Understanding these networks' structure and features and locating significant actors, influential figures, and interaction patterns are the main objectives of SNA. SNA often entails gathering information on the connections between people, visualizing these connections as a network graph, and then analyzing the resulting graph using mathematical and statistical methods. To better understand the dynamics of social systems and to guide decision-making, the insights gleaned by SNA can be applied to several applications, including behavior, public health, organizational behavior, and political science.
Zachary collected information about the original club and the dataset obtained is now known as “Zachary’s karate club” network. The groups that emerged from the fission of the karate club were factions not necessarily recognised by the club members. Instead, the friendship network among members gave rise to them during a moment of conflict. The dataset called karate.gml can be obtained at https://doi.org/10.6084/m9.figshare.7985174.v116 and 16 Rogel-Salazar, J. (2019c, Apr). Zachary’s karate club. https://doi.org/10.6084/ m9.figshare.7985174.v1 we will use it in the rest of this section. It contains 34 nodes representing individuals within the karate club. The edges in the network are given by interactions between two individuals outside the activities of the club such as actual lessons or meetings.
The project makes use of NetwokX library for analysis and concluding about the clusters,centrality measures and insights of the social network.
The analysis gives the result of the split of the Karate club due to a few issues.John A ans Hi are the two prominant nodes whose behaviour and analysis must me studied.
Output screenshots are uploaded.
1.degree centrality
For finding very connected individuals, popular individuals, individuals who are likely to hold most information or individuals who can quickly connect with the wider network.
2.betweeness centrality
For finding the individuals who influence the flow around a system.
3.closeness centrality
For finding the individuals who are best placed to influence the entire network most quickly.
4.eigen value centrality
EigenCentrality is a good ‘all-round’ SNA score, handy for understanding human social networks, but also for understanding networks like malware propagation.
5.PageRank centrality
Because it takes into account direction and connection weight, PageRank can be helpful for understanding citations and authority.