GRAPH NEURAL NETWORKS FOR SOCIAL NETWORK ANALYSISIN INDIA: DETECTING FAKE PROFILES & BOTNETS

Abstract

Graph Neural Networks (GNNs) have emerged as a powerful approach for analyzing complex social network structures and identifying malicious activities such as fake profiles and botnets. This study presents a GNN-based framework for social network analysis in India, focusing on the detection of fake accounts, automated bots, and coordinated malicious behavior across online platforms. The proposed system utilizes user interaction patterns, profile attributes, connectivity structures, and behavioral features to construct graph-based representations of social media networks. Advanced GNN architectures such as Graph Convolutional Networks (GCN) and Graph Attention Networks (GAT) are employed to learn hidden relationships and classify suspicious nodes with high accuracy. The model enhances cybersecurity by detecting anomalous communication patterns, reducing misinformation spread, and improving trust within digital communities. Experimental results demonstrate that the proposed approach outperforms traditional machine learning methods in terms of detection accuracy, scalability, and adaptability to evolving bot behaviors. This research highlights the significance of graph-based deep learning techniques in strengthening social media security and protecting online users in the rapidly growing Indian digital ecosystem.