Analyzing Interest-Based Homophily in Online Social Networks Using Community Detection Methods

Abstract

A social network is a structure made up of individuals, groups, or organizations connected by relationships. These relationships can be friendships, family bonds, or professional links. In this study, the focus is on homophily, which means people with similar interests tend to connect. The research analyzes a dataset of 1,000 users from [social media platform] to understand how shared interests affect community formation. Pearson correlation measures interest similarity between users. The Label Propagation Algorithm (LPA) and the Louvain method help detect user communities. The study finds distinct communities, where users are grouped based on common interests. The results show that homophily strongly influences how online communities form. This research provides a simple method to analyze user connections and improve community detection in social networks.