Exploring Link Prediction Techniques in Social Network Analysis for Community Detection

The global proliferation of social media networks has sparked significant attention, with platforms such as Facebook, Twitter, Telegram, and WhatsApp enabling widespread interaction among users. Consequently, addressing the challenges associated with social networking sites has become crucial. To tackle this issue, a recommendation model has been developed, merging individuals with similar interests through a community detection algorithm. This statistical approach mines the dynamic social communities based on comment patterns. The model addresses the pressing concern of identifying nodes with a high prevalence of combining based on the interest. The integrated model incorporates link prediction and sentiment analysis features, facilitating the creation of communities, and thereby allowing the identification of individuals with shared interests within a single cluster. To evaluate the performance of the model, a dataset of 5672 Facebook page comments was collected from the Stanford repository. The result shows using the edge-betweeness technique of the community detection algorithm maximum modularity density of 0.2331 is obtained.