Enhancing Social Media User's Trust: A Comprehensive Framework for Detecting Malicious Profiles Using Multi-Dimensional Analytics

Abstract

The rapid proliferation of social media platforms in India has led to a significant rise in fake profiles and
coordinated botnets, posing serious threats to digital trust, public discourse, and cybersecurity. Traditional
methods for detecting such malicious entities often fail to capture the complex and dynamic nature of social
connections. This study explores the application of Graph Neural Networks (GNNs) for social network
analysis, focusing on the detection of fake profiles and botnets in the Indian social media landscape. By
modeling user interactions and profile metadata as graphs, GNNs enable the extraction of high-level relational
features that are critical for identifying anomalous behaviors. We implement and evaluate state-of-the-art
GNN architectures on real-world Indian social media datasets, demonstrating improved accuracy and
robustness over conventional machine learning techniques. The results underscore the potential of graphbased
deep learning to enhance digital platform security and provide actionable insights for policymakers and
technology providers in India