A Hybrid Algorithm for Scalable Friend Recommendation in Large-Scale Social Networks

The rapid growth of online social networks has led to a surge in user-generated data, resulting in significant information overload and making accurate friend recommendations increasingly difficult. Traditional recommendation methods—such as social graph analysis, collaborative filtering, and semantic similarity—struggle to scale effectively in large-scale environments due to high computational costs or limited precision. This research proposes a hybrid algorithm that combines location-based proximity detection and user classification to enhance the accuracy and scalability of friend recommendations. By identifying geographically or behaviorally close user clusters and separating distinct user classes, the system can efficiently uncover potential social connections. Experimental results demonstrate that the proposed method improves friend suggestion relevance while reducing computational overhead, making it suitable for large and dynamic social platforms.