SNITCH: Smart Network-based Intelligent Threat Classifier for Phishing Detection using SVM Active Learning

DHAYANITHISELVAN, M.; JOSHWAA, A. PONEY; PRASATH, S.K. JAYA

doi:https://doi.org/10.55041/ijcope.v2i5.798

Published on: May 2026

SNITCH: SMART NETWORK-BASED INTELLIGENT THREAT CLASSIFIER FOR PHISHING DETECTION USING SVM ACTIVE LEARNING

M. DHAYANITHISELVAN A. PONEY JOSHWAA S.K. JAYA PRASATH

N.KANAGADURGA

Department of Computer Science and Engineering, E.G.S. Pillay Engineering College (Autonomous), Nagapattinam, Tamilnadu, India

DOI:https://doi.org/10.55041/ijcope.v2i5.798

Article Status

Plagiarism Passed Peer Reviewed Open Access

Available Documents

Download PDF Review Report

Abstract

SNITCH (Smart Network-based Intelligent Threat Classifier for Phishing Detection using SVM Active Learning) is an intelligent phishing URL detection framework that combines Support Vector Machine (SVM) classification with an uncertainty-driven Active Learning strategy. Phishing attacks remain one of the most prevalent cybersecurity threats, targeting users through deceptive URLs to steal sensitive information such as banking credentials, passwords, and personal data. Conventional blacklist and rule-based detection approaches are inherently reactive and fail to identify zero-day phishing attempts. SNITCH employs an SVM classifier with an RBF kernel, trained on a curated dataset of over 48,000 URLs drawn from PhishTank and the UCI Machine Learning Repository. More than 22 lexical, structural, and domain-based features are extracted from each URL. Class imbalance was resolved using SMOTE, and hyperparameters were optimized through GridSearchCV with 5-fold cross-validation, yielding C = 10 and Gamma = 0.01. The Active Learning pipeline employs uncertainty sampling to select only the most informative samples for labeling, reducing annotation requirements by approximately 40% compared to passive learning. SNITCH achieved 87% accuracy, 86% precision, 85% recall, an F1-score of 85.5%, and a ROC-AUC of 0.89, outperforming the passive learning baseline across all metrics.

Keywords — Phishing Detection, Support Vector Machine, Active Learning, Uncertainty Sampling, SMOTE, URL Feature Extraction, Cybersecurity, Machine Learning

How to Cite this Paper

DHAYANITHISELVAN, M., JOSHWAA, A. P. & PRASATH, S. J. (2026). SNITCH: Smart Network-based Intelligent Threat Classifier for Phishing Detection using SVM Active Learning. International Journal of Creative and Open Research in Engineering and Management, <i>02</i>(05). https://doi.org/10.55041/ijcope.v2i5.798

DHAYANITHISELVAN, M., et al.. "SNITCH: Smart Network-based Intelligent Threat Classifier for Phishing Detection using SVM Active Learning." International Journal of Creative and Open Research in Engineering and Management, vol. 02, no. 05, 2026, pp. . doi:https://doi.org/10.55041/ijcope.v2i5.798.

DHAYANITHISELVAN, M.,A. JOSHWAA, and S.K. PRASATH. "SNITCH: Smart Network-based Intelligent Threat Classifier for Phishing Detection using SVM Active Learning." International Journal of Creative and Open Research in Engineering and Management 02, no. 05 (2026). https://doi.org/https://doi.org/10.55041/ijcope.v2i5.798.

Search & Index

References

[1] P. Singh and R. Kumar, "URL-Based Phishing Detection Using Machine Learning Techniques," International Journal of Cybersecurity Intelligence, vol. 8, no. 2, pp. 45–61, 2024.

[2] N. Abdelhamid, C. Ait Aouiti, and H. Kheddouci, "Phishing Website Detection Using Support Vector Machine," Journal of Information Security and Applications, vol. 15, no. 3, pp. 112–128, 2024.

[3] B. Settles, "Active Learning Literature Survey," University of Wisconsin Technical Report, vol. 1648, pp. 1–67, 2024.

[4] R. Verma and A. Das, "Deep Learning Approaches for Phishing URL Detection Using Character-Level LSTM," in Proc. IEEE International Conference on Cybersecurity, pp. 230–245, 2023.

[5] C. Fernandez and M. Santos, "SMOTE-Enhanced SVM for Imbalanced Cybersecurity Classification," Journal of Machine Learning Applications, vol. 5, no. 1, pp. 78–93, 2023.

[6] F. Pedregosa et al., "Scikit-Learn: Machine Learning in Python," Journal of Machine Learning Research, vol. 12, pp. 2825–2830, 2021.

[7] N. V. Chawla, K. W. Bowyer, L. O. Hall, and W. P. Kegelmeyer, "SMOTE: Synthetic Minority Over-sampling Technique," Journal of Artificial Intelligence Research, vol. 16, pp. 321–357, 2020.

[8] C. Cortes and V. Vapnik, "Support-Vector Networks," Machine Learning Journal, vol. 20, no. 3, pp. 273–297, 2019.

[9] D. Sahoo, C. Liu, and S. C. H. Hoi, "Malicious URL Detection Using Machine Learning: A Survey," ACM Computing Surveys, vol. 54, no. 1, pp. 1–35, 2021.

[10] J. Ma, L. K. Saul, S. Savage, and G. M. Voelker, "Beyond Blacklists: Learning to Detect Malicious Web Sites from Suspicious URLs," in Proc. ACM SIGKDD, pp. 1245–1253, 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: May 28 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 05 View All Issues Next Article

← Previous Article

SmartShop: A Local LLM-Based Conversational Kiosk for Product Navigation and Sales Assistance in Supermarkets

Next Article →

Social Media Brand Modeling: A Predictive Study on Consumer Engagement