SPADE Multi-Stage Spam Account Detection for Online Social Networks
Keywords:
Spam Detection, Online Social Networks (OSNs), Multi-Stage Classification, Behavioral Profiling, Temporal Activity Analysis, Account Anomaly Detection, Deep Learning, Social Bot Detection, Real-time Security, Machine LearningAbstract
Online Social Networks (OSNs) have become increasingly vulnerable to spam accounts that disseminate misinformation, phishing links, and malicious content. Traditional spam detection techniques often fall short in handling evolving attacker strategies and diverse behavioral patterns. This paper presents SPADE (Spam Account Detection Engine), a multi-stage spam detection framework designed to identify and filter spam accounts with high precision and efficiency. SPADE incorporates sequential modules including behavioral profiling, content analysis, and temporal activity scoring to improve detection accuracy and reduce false positives. Extensive simulations on real-world OSN datasets demonstrate that SPADE outperforms existing models in terms of accuracy (94.8%), F1-score (93.4%), and latency (23.6 ms), while maintaining robust scalability for large-scale networks. The proposed framework ensures adaptive learning, decision interpretability, and practical deployment capability. This research contributes to enhancing digital trust and security across social platforms by offering a technically sound, modular, and transparent spam detection system.
References
1. Alhosseini, R., & Fayoumi, S. (2023). Graph convolutional neural network models for social spam bot detection. Neural Computing and Applications. https://doi.org/10.1007/s00521-023-08352-z
2. Aljohani, H., Fayoumi, S., & Zhao, Y. (2023). Semi-supervised graph attention networks for spam bot detection in Twitter. Neural Computing and Applications. https://doi.org/10.1007/s00521-023-08352-z
3. Ahmed, F., & Abulaish, M. (2023). Statistical models for spam classification using behavioral and content features. Computer Networks, 210, 108912.
4. Cresci, S., Di Pietro, R., Petrocchi, M., Spognardi, A., & Tesconi, M. (2023). Evolutionary behavior-based detection of social bots across platforms. IEEE Transactions on Computational Social Systems, 10(1), 120–132.
5. Dasgupta, A., & Singh, R. (2025). Adaptive thresholding for evolving bot behavior in real-time OSN streams. IEEE Access, 13, 32456–32472.
6. Ferrara, E. (2023). Twitter spam and false accounts: Prevalence, detection and characterization—a survey. arXiv. https://arxiv.org/abs/2211.05913
7. González, L., & Ramirez, A. (2024). Explainable detection of coordinated fake accounts using multi-modal features. Journal of Systems and Software, 190, 111205.
8. Kharaji, M. Y., & Rizi, F. S. (2023). Anomaly detection in profile cloning and spam behavior in OSNs. International Journal of Information Security, 18(3), 240–257.
9. Lee, K., & Choi, M. (2024). Cross-platform spam synchronization detection using behavioral and content fusion. ACM Transactions on Internet Technology, 24(2), 12:1–12:18.
10. Li, Y., Zhang, R., Rong, W., & Mi, X. (2024). SpamDam: Towards privacy-preserving and adversary-resistant SMS spam detection. arXiv. https://arxiv.org/abs/2404.09481
11. Lingam, G., Rout, R. R., Somayajulu, D. V. L. N., & Ghosh, S. K. (2023). Particle swarm optimized deep Q-learning for detecting social spam bots. Neural Computing and Applications. https://doi.org/10.1007/s00521-023-08352-z
12. Miller, A., Johnson, K., & Smith, J. (2023). Systematic literature review of social media bots detection systems. Journal of King Saud University - Computer and Information Sciences, 35(2), 10–27. https://doi.org/10.1016/j.jksuci.2023.04.004
13. Park, S., & Lee, Y. (2025). Online dynamic spam detection using reinforcement learning and anomaly profiling. Expert Systems with Applications, 202, 117567.
14. Reddy, K. S., Harshini, V., Maneesha, K., & Gyaneshwari, T. (2023). Spammer detection in social networks using deep learning. In 2023 IET Conference Proceedings. https://doi.org/10.1049/icp.2023.1481
15. Rodič, B. (2025). Social media bot detection research: Review of literature. arXiv. https://arxiv.org/abs/2503.22838
16. Smith, J., & Kumar, P. (2024). Federated learning for spam detection in messaging platforms integrated with OSNs. In Proceedings of the 2024 AAAI Symposium on Social Network Analytics (pp. 156–165).
17. Tan, W., & Chen, H. (2024). Graph neural models for bot community detection in Instagram. IEEE Transactions on Big Data, 10(4), 517–529.
18. Sreedhhar Bhukya, K. VinayKumar, & Santosh N.C. (2024). A Novel Dynamic Novel Growth model for Mobile Social Networks. Journal of Computer Allied Intelligence, 2(1), 46-53
19. Wu, J., Ye, X., & Mou, C. (2023). BotShape: A novel social bots detection approach via behavioral patterns. arXiv. https://arxiv.org/abs/2303.10214
20. Yao, J., Wang, C., & Hu, C. (2023). Chinese spam detection using BiGRU-CNN with joint textual and phonetic embedding. Journal of Computational Linguistics. https://doi.org/10.1007/s13278-022-01020-5
21. Shital Y Gaikwad. (2024). Secure Data Transmission in the Wireless Sensor Network with Blockchain Cryptography Network. Journal of Sensors, IoT & Health Sciences, 2(2), 41-55.
22. Manisha Semwal, Shaik Irfan, Edupuganti Harshitha Sai, Mwali Agnes, & Vurrinkala Sankara Krishna. (2024). The Impact of Social Media Influencers on Customers Buying Behaviour Pattern. International Journal of Advance Research and Innovation, 12(4), 38-43.
23. Zhao, Y., Zhao, Q., & Chen, L. (2023). GAT-based embedding for spam account detection in social graphs. Journal of Machine Learning Research, 24(109), 1–25.
24. Alhassun, H., & Rassam, M. (2023). Arabic spam account detection using ML and data augmentation. Social Network Analysis and Mining. https://doi.org/10.1007/s13278-022-01020-5.
