Abstract :

The Rapid Growth Of Social Networking Platforms In India Has Led To A Significant Rise In Fake Profiles, Automated Bot Accounts, And Coordinated Misinformation Campaigns. These Malicious Entities Pose Serious Threats To User Privacy, Public Opinion, Digital Trust, And National Security. Traditional Detection Techniques—such As Rule-based Filtering, Heuristic Analysis, And Classical Machine Learning Models—struggle To Accurately Identify Sophisticated Fake Accounts Due To Evolving Behavioral Patterns And The Complex Relational Structures Within Social Networks. To Address These Challenges, This Study Proposes A Graph Neural Network (GNN)-based Framework For Detecting Fake Profiles And Botnets In Indian Social Media Ecosystems. The Proposed Model Leverages The Graph-like Structure Of Social Networks, Where Nodes Represent User Accounts And Edges Represent Interactions Such As Likes, Comments, Messages, And Follows. Using GNN Architectures Such As Graph Convolutional Networks (GCN), Graph Attention Networks (GAT), And GraphSAGE, The System Learns Both Structural And Behavioral Embeddings To Distinguish Between Genuine And Fraudulent Accounts. Key Features—including Activity Frequency, Interaction Diversity, Temporal Patterns, Linguistic Cues, And Follower-following Relationships—are Integrated To Improve Detection Accuracy. Experimental Evaluation On Real-world And Synthetic Datasets Demonstrates That GNN-based Models Outperform Traditional Machine Learning Classifiers In Identifying Bot Clusters, Anomalous Communication Patterns, And Fake-profile Communities. The Framework Achieves Higher Precision, Recall, And F1-scores, Especially In Highly Imbalanced Datasets Common In Fraud Detection Tasks. The System’s Scalability And Adaptability Make It Suitable For Large Indian Platforms Such As ShareChat, Moj, And Region-specific Social Networks. Overall, This Research Highlights The Effectiveness Of Graph Neural Networks In Combating Online Manipulation By Uncovering Hidden Patterns And Relational Anomalies Within Social Graphs. The Proposed Approach Provides A Robust, Intelligent Solution For Enhancing Digital Safety And Maintaining User Trust In Indian Social Media Environments. KEYWORDS: Graph Neural Networks (GNN), Fake Profile Detection, Botnet Detection, Social Network Analysis, Graph Convolutional Networks (GCN), Graph Attention Networks (GAT), GraphSAGE, Anomaly Detection, User BehaviorModeling, Indian Social Networks, Deep Learning, Misinformation Detection.

Published:

18-12-2025

Issue:

Vol. 25 No. 12 (2025)

Page Nos:

327-337

Section:

Articles

License:

This work is licensed under a Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 International License.

How to Cite