Klastry węzłów i podsieci

In Social Network Analysis (SNA), one of the key issues is to understand the structure of the network by identifying different groups (communities) within a larger set of nodes. After discussing concepts such as centrality (the importance of a node in the network) and structural holes (places that can be bridges between different groups), we move on to the analysis of clusters and subnetworks. Both concepts allow us to better understand how networks organize themselves into subgroups and how these groups affect the entire system.

A cluster is a group of nodes that are densely connected to each other, but less connected to other nodes in the network. What distinguishes clusters is the internal density of connections and the differences in connections to nodes outside the cluster. This phenomenon is common in social networks, where we can see the emergence of smaller communities, such as groups of friends on social media, work teams in companies, or interest groups.

A subnet is any distinct part of a network that consists of a subset of nodes and the connections between them. A subnet can include nodes from different parts of the network, and its selection can be made based on various criteria, such as topic, time, location, or type of relationship.

Clusters in Social Network Analysis

Clusters are not only an interesting theoretical element, but also have a wide practical application in the analysis of social networks. Identifying clusters allows for:

• Community detection: In social networks, clusters can represent groups that are more closely related to each other, such as Facebook friends, coworkers at a company, or participants in a common hobby.
• Information bubble analysis: People within a cluster can only share similar opinions and information, which leads to the creation of “information bubbles” in the network.
• Identifying leaders: Within a cluster, there are often individuals with high centrality who act as opinion leaders or mediators between group members.