Resilient Semantic Threat Detection at the Edge: A Knowledge Distillation Framework for SMS Spam Classification

Mrinal,; Kumar, Neeraj

doi:https://doi.org/10.55041/ijcope.v2i8.090

Volume 02, Issue 8

Published on: August 2026

RESILIENT SEMANTIC THREAT DETECTION AT THE EDGE: A KNOWLEDGE DISTILLATION FRAMEWORK FOR SMS SPAM CLASSIFICATION

Mrinal Neeraj Kumar

Department Of Computer Science And Engineering,

MERI College Of Engineering And Technology, Haryana, India.

DOI:https://doi.org/10.55041/ijcope.v2i8.090

Article Status

Plagiarism Passed Peer Reviewed Open Access

Available Documents

Download PDF Review Report

Abstract

The ubiquity of Application-to-Person (A2P) messaging has inadvertently created a robust vector for mobile security threats, specifically Smishing (SMS Phishing). Traditional lexical filters, such as Naive Bayes and Support Vector Machines (SVM), exhibit diminishing returns against modern obfuscated attacks due to their inability to interpret semantic context. While Transformer-based architectures like BERT have revolutionized Natural Language Processing (NLP), their computational latency renders them impractical for real-time deployment on resource-constrained mobile edge devices. This study proposes a high-efficiency detection framework utilizing DistilBERT, a distilled knowledge representation of the BERT transformer. By fine-tuning this architecture on the UCI SMS Spam Collection, we achieved a testing accuracy of 99.04% and a weighted F1-score of 0.9904. Notably, the model maintained a precision of 0.97 on the minority spam class, effectively mitigating the class imbalance problem without synthetic data augmentation. These results substantiate the viability of Knowledge Distillation as a mechanism to deploy state-of-the-art semantic security filters on edge infrastructure.

Keywords - SMS Spam, Knowledge Distillation, DistilBERT, Cybersecurity, Edge AI, NLP, Smishing, Transfer Learning

How to Cite this Paper

Mrinal, & Kumar, N. (2026). Resilient Semantic Threat Detection at the Edge: A Knowledge Distillation Framework for SMS Spam Classification. International Journal of Creative and Open Research in Engineering and Management, <i>02</i>(8), 1-9. https://doi.org/10.55041/ijcope.v2i8.090

Mrinal, , and Neeraj Kumar. "Resilient Semantic Threat Detection at the Edge: A Knowledge Distillation Framework for SMS Spam Classification." International Journal of Creative and Open Research in Engineering and Management, vol. 02, no. 8, 2026, pp. 1-9. doi:https://doi.org/10.55041/ijcope.v2i8.090.

Mrinal, , and Neeraj Kumar. "Resilient Semantic Threat Detection at the Edge: A Knowledge Distillation Framework for SMS Spam Classification." International Journal of Creative and Open Research in Engineering and Management 02, no. 8 (2026): 1-9. https://doi.org/https://doi.org/10.55041/ijcope.v2i8.090.

Search & Index

References

Gupta and A. Kumar, “A Comprehensive Survey on Smishing Attacks and Defense Mechanisms in 5G Networks,” IEEE Access, vol. 11, pp. 15678–15695, 2023.

Smith et al., “The State of Mobile Security 2024: Trends and Predictions,” Computers & Security, vol. 132, p. 103345, 2024.

Global Cybersecurity Forum, “Annual Report on Digital Fraud and Phishing 2023,” GCF Reports,

Zhang, Y. Wang, and X. Chen, “Adversarial Text Generation for Evaluating SMS Spam Detectors,” Proc. ACM SIGSAC, 2023.

Al-Fayoumi et al., “Evasion Techniques in SMS Phishing: A 2024 Perspective,” Int. J. Inf. Secur., vol. 23, no. 2, pp. 201–215, 2024.

K. Singh and P. K. Singh, “Handling Class Imbalance in Short Text Classification: A Review,” IEEE Trans. Knowl. Data Eng., vol. 34, no. 8, pp. 3891–3908, 2022.

Ghourabi, “A Hybrid CNN-LSTM Model for SMS Spam Detection,” Int. J. Interact. Mob. Technol., vol. 14, no. 13, 2020.

Liu et al., “Efficient Transformers for Mobile Edge Computing: A Survey,” IEEE Internet Things J., vol. 9, no. 18, pp. 17256–17274, 2022.

Sanh et al., “DistilBERT, a distilled version of BERT: smaller, faster, cheaper and lighter,” arXiv preprint arXiv:1910.01108, [Cited as seminal architecture reference, widely used in 2020+ research].

Wolf et al., “Transformers: State-of-the-art natural language processing,” Proc. EMNLP, 2020.

Ethical Compliance & Review Process

•All submissions are screened under plagiarism detection.
•Review follows editorial policy.
•Authors retain copyright.
•Peer Review Type: Double-Blind Peer Review
•Published on: Aug 11 2026

CCBYNC

This article is licensed under a Creative Commons Attribution-NonCommercial 4.0 International License. You are free to share and adapt this work for non-commercial purposes with proper attribution.

View License

Back to Volume 02, Issue 8 View All Issues Next Article

← Previous Article

Quantum-Resonant Energy Coupling–Molecular Resonance Technology: A Quantitative and Falsifiable Multidomain Validation Framework with an Exploratory Coal Case Study

Next Article →

Revisiting the Bhagavad Gita through the Indian Knowledge System: Implications for Hospitality Management