The project leverages the R libraries such as igraph, network, sna, and visNetwork to perform Social Network Analysis (SNA). It visualizes relationships between nodes (vertices) in a network, calculates various centrality measures like degree, closeness, and betweenness, and highlights hubs, authorities, and communities within the network.
The visual representation of these graphs is enhanced using customization features, allowing for color changes to vertices and edges, size adjustments, and layout manipulation. The project moves from a simple network of numerical nodes to a named network of entities (e.g., “Simon”, “Josh”, etc.), displaying their interconnections in visual plots. Additionally, centrality measures such as degree, closeness, betweenness, and edge betweenness are calculated, helping to identify the most important or influential nodes in the network. For example, degree(g1) gives the number of connections each node has, and betweenness(g1) identifies the nodes that frequently appear on the shortest paths between other nodes.
This Social Network Analysis project in R utilizes multiple libraries like igraph, network, sna, and visNetwork to analyze the relationships between entities (nodes) in a network. The project starts by installing these packages and loading the igraph library, which serves as the core tool for constructing and analyzing network graphs. The network graphs are created using the graph() function, which connects vertices (nodes) with edges (links). Various types of networks are explored, including simple graphs and undirected graphs.
The project also includes a real-world application where CSV data is imported to form a network graph using the graph.data.frame() function. Node degrees are visualized in histograms, highlighting the distribution of connections across the network. More advanced concepts like hubs and authorities are explored using hub_score() and authority_score(), respectively, demonstrating the influential players in the network.
Finally, community detection techniques such as edge betweenness clustering are applied to identify groups of nodes that are more tightly connected to each other than to the rest of the network. This is illustrated through visual plots, enabling a clearer understanding of the social structures within the dataset. Overall, the project provides a comprehensive approach to analyzing and visualizing social networks, with applications in fields like sociology, biology, and online networks.