Mutual Clustering Coefficient-Based Suspicious-Link Detection Enhancing Sentiment Analysis in Social Networks

The exponential growth of social networks has led to the proliferation of misleading, malicious, or suspicious content, significantly impacting user interactions and societal perceptions. To address these challenges, this study proposes a Mutual Clustering Coefficient (MCC) based Suspicious-Link Detection model, which enhances sentiment analysis in social networks. The model leverages MCC to detect suspicious links that could influence the spread of false or harmful information. By analyzing the relationships between users and the density of connections within clusters, the model identifies anomalies that deviate from normal user interactions, thereby flagging suspicious links. Additionally, the model incorporates sentiment analysis to evaluate the emotional tone of interactions surrounding these links. This dual approach analyzing both network structure and sentiment provides a comprehensive method for early detection of harmful content. The proposed system is validated using large-scale Online Social Network (OSN) datasets, demonstrating high accuracy in detecting suspicious links while enhancing the precision of sentiment analysis. By integrating both MCC and sentiment analysis, this research aims to improve the safety, trustworthiness, and overall health of social network platforms,