International Journal of Innovative Research in Computer and Communication Engineering
ISSN Approved Journal | Impact factor: 8.771 | ESTD: 2013 | Follows UGC CARE Journal Norms and Guidelines
| Monthly, Peer-Reviewed, Refereed, Scholarly, Multidisciplinary and Open Access Journal | High Impact Factor 8.771 (Calculated by Google Scholar and Semantic Scholar | AI-Powered Research Tool | Indexing in all Major Database & Metadata, Citation Generator | Digital Object Identifier (DOI) |
| TITLE | Fake Account Detection (Risk Analysis) in Social Media using GAN |
|---|---|
| ABSTRACT | The proliferation of automated and artificially-generated accounts on social media platforms, particularly Twitter/X, has emerged as a critical threat to digital ecosystem integrity, information veracity, and public discourse authenticity. Traditional rule-based detection systems have become ineffective against evolving bot strategies that employ sophisticated behavioral mimicry, AI-generated content, and coordinated network attacks. This research presents a comprehensive Fake Profile Detection System that integrates metadata analysis, behavioral pattern recognition, content-based features, network topology inspection, and adversarial verification mechanisms. The proposed hybrid framework employs gradient-boosted decision tree classification combined with a deep-learning-based discriminator model trained on both real and synthetic profile data. Our system achieves 96% detection accuracy while maintaining low false-positive rates, demonstrating substantial improvements over single-feature and rule-based baseline approaches. The architecture emphasizes scalability, real-time processing capability, and modular design for seamless deployment in production environments. We further incorporate ethical safeguards including privacy preservation, transparent decision-making, and data anonymization protocols. The system effectively identifies both elementary bots and sophisticated AI-driven fake profiles across diverse user populations, addressing critical gaps in existing detection methodologies. |
| TITLE | |
| AUTHOR | ANU C S, DISHA V K, JAHNAVI M MALLADAD, KARTHIK M DAIVADNYA, BHARATH V Assistant Professor, Department of CS&E, Bapuji Institute of Engineering and Technology, Davanagere, Karnataka, India U.G. Student, Department of CS&E, Bapuji Institute of Engineering and Technology, Davanagere, Karnataka, India |
| VOLUME | 177 |
| DOI | DOI: 10.15680/IJIRCCE.2025.1312140 |
| pdf/140_Fake Account Detection (Risk Analysis) in Social Media.pdf | |
| KEYWORDS | |
| References | [1] Sharma, R., & Gupta, A. (2024). Adversarial bot detection: Using GANs to synthesize fake profiles for robust classification. IEEE Transactions on Computational Social Systems, 11(3), 455–468. [2] Khan, M., Verma, S., & Al-Taher, O. (2024). Hybrid model combining content, behavior, and network features for social bot detection. ACM Transactions on the Web, 18(2), 210–225. [3] Iyer, P., & Das, R. (2024). Explainable machine learning for bot detection: Feature importance and model transparency. Journal of Information and Data Science, 9(1), 55–70. [4] Liu, Y., Park, J., & Santos, M. (2024). Adversarial evasion resistance in bot detection systems. IEEE Access, 12, 22850–22863. [5] Fernando, J., & Lee, H. (2024). Scaling fake profile detection to millions of users: A big data approach. Journal of Big Data Analytics, 7(4), 301–320. Wong, M., & Ibrahim, R. (2023). Lightweight fake profile detection for mobile applications using minimal metadata features. Mobile Information Systems, 2023, 1–14. [6] Cheng, L., & Roy, S. (2023). Lightweight fake profile detection for mobile applications. Mobile Computing and Applications Review, 5(2), 89–102. [7] Martínez, A., Kumari, D., & Zhou, T. (2023). Cross-language fake profile detection on multilingual social media. International Journal of Multilingual Data Mining, 12(1), 44–58. [8] Hassan, F., & Kim, D. (2023). Temporal pattern analysis of tweet activity for bot detection. Social Network Analytics Journal, 9(3), 112–128.Zhang, W., & Kim, H. (2023). Real-time fake account detection using streaming Twitter data and lightweight gradient boosting models. IEEE Access, 11, 47620–47634. [9] Patel, S., & Narang, V. (2023). Combined metadata and behavioral feature learning for social bot detection. Expert Systems and Applications, 211, 119–132. [10] O’Connor, B., & Singh, R. (2023). Real-time fake account detection using streaming data from social platforms. IEEE Internet Computing, 27(5), 37–48. |
