Optimasi Hybrid SVM-Naïve Bayes, Weighted Voting Deteksi Phishing URL, Seleksi Fitur-Explainable AI
Keywords:
Phishing URL, Support Vector Machine (SVM), Naïve Bayes, Feature Selection, Explainable AIAbstract
Serangan phishing merupakan ancaman besar dalam keamanan siber yang memanfaatkan rekayasa sosial untuk memperoleh informasi sensitif seperti kredensial login atau data perbankan. Salah satu upaya penanggulangannya yang efektif adalah deteksi URL phishing berbasis machine learning, yang dapat mengidentifikasi pola berbahaya secara otomatis. Penelitian ini mengusulkan model hybrid dengan menggabungkan algoritma Support Vector Machine (SVM) dan Naïve Bayes melalui pendekatan Weighted Voting untuk meningkatkan akurasi deteksi dan kestabilan model. Untuk mengurangi overfitting dan meningkatkan efisiensi klasifikasi, dilakukan seleksi fitur menggunakan SelectKBest, Recursive Feature Elimination (RFE), dan Principal Component Analysis (PCA). Ketiga metode dibandingkan untuk menilai kontribusinya terhadap performa model. Dalam rangka meningkatkan transparansi dan interpretabilitas, pendekatan Explainable AI (XAI) seperti SHAP dan LIME digunakan untuk menganalisis kontribusi fitur terhadap hasil prediksi. Performa model dievaluasi berdasarkan metrik akurasi, presisi, recall, dan F1-score. Hasil eksperimen menunjukkan bahwa model hybrid dengan fitur optimal mampu mencapai akurasi tinggi dan menghasilkan interpretasi prediktif yang informatif. Kombinasi pendekatan algoritmik dan interpretatif ini menawarkan solusi yang akurat sekaligus dapat dijelaskan, sehingga cocok diterapkan dalam sistem deteksi phishing yang efektif. Temuan ini diharapkan berkontribusi pada pengembangan sistem keamanan berbasis machine learning yang transparan dan efisien untuk menghadapi ancaman phishing di dunia nyata.
References
[1] A. F. Mahmud and S. Wirawan, “Deteksi Phishing Website menggunakan Machine Learning Metode Klasifikasi,” Sistemasi: Jurnal Sistem Informasi, vol. 13, no. 4, pp. 1368–1380, 2024.
[2] A. Mutemi and F. Bacao, “E-commerce fraud detection based on machine learning techniques: Systematic literature review,” Big Data Mining and Analytics, vol. 7, no. 2, pp. 419–444, 2024.
[3] A. Safi and S. Singh, “A systematic literature review on phishing website detection techniques,” Journal of King Saud University-Computer and Information Sciences, vol. 35, no. 2, pp. 590–611, 2023.
[4] A. F. Nugraha, R. F. A. Aziza, and Y. Pristyanto, “Penerapan metode Stacking dan Random Forest untuk Meningkatkan Kinerja Klasifikasi pada Proses Deteksi Web Phishing,” Jurnal Infomedia: Teknik Informatika, Multimedia, dan Jaringan, vol. 7, no. 1, pp. 39–44, 2022.
[5] D. C. Asogwa, S. O. Anigbogu, I. E. Onyenwe, and F. A. Sani, “Text classification using hybrid machine learning algorithms on big data,” arXiv preprint arXiv:2103.16624, 2021.
[6] P. P. Kuantitatif, “Metode Penelitian Kunatitatif Kualitatif dan R&D,” Alfabeta, Bandung, 2016.
[7] N. Gharaibeh, O. M. Al-hazaimeh, A. Abu-Ein, and K. M. Nahar, “A hybrid svm naïve-bayes classifier for bright lesions recognition in eye fundus images,” Int J Electr Eng Inform, vol. 13, no. 3, pp. 530–545, 2021.
[8] L. C. Nnadi, Y. Watanobe, M. M. Rahman, and A. M. John-Otumu, “Prediction of students’ adaptability using explainable AI in educational machine learning models,” Applied Sciences, vol. 14, no. 12, p. 5141, 2024.
[9] S. R. Abdul Samad et al., “Analysis of the performance impact of fine-tuned machine learning model for phishing URL detection,” Electronics (Basel), vol. 12, no. 7, p. 1642, 2023.
[10] S. Wang, J. Ren, and R. Bai, “A semi-supervised adaptive discriminative discretization method improving discrimination power of regularized naive Bayes,” Expert Syst Appl, vol. 225, p. 120094, 2023.
[11] Z. Alhaq, A. Mustopa, S. Mulyatun, and J. D. Santoso, “Penerapan Metode Support Vector Machine Untuk Analisis Sentimen Pengguna Twitter,” Journal of Information System Management (JOISM), vol. 3, no. 1, pp. 16–21, 2021.
[12] A. Rojarath and W. Songpan, “Cost-sensitive probability for weighted voting in an ensemble model for multi-class classification problems,” Applied Intelligence, vol. 51, pp. 4908–4932, 2021.
[13] V. C. Osamor and A. F. Okezie, “Enhancing the weighted voting ensemble algorithm for tuberculosis predictive diagnosis,” Sci Rep, vol. 11, no. 1, p. 14806, 2021.
[14] W. Yang et al., “Survey on explainable AI: From approaches, limitations and applications aspects,” Human-Centric Intelligent Systems, vol. 3, no. 3, pp. 161–188, 2023.
[15] V. Vimbi, N. Shaffi, and M. Mahmud, “Interpreting artificial intelligence models: a systematic review on the application of LIME and SHAP in Alzheimer’s disease detection,” Brain Inform, vol. 11, no. 1, p. 10, 2024.
Downloads
Published
Issue
Section
License
All articles published in the eProsiding Sistem Informasi (POTENSI) are licensed under the provisions of the Study Program. Under this license, authors retain the copyright to their works, while others are permitted to copy, distribute, display, and create derivative works from the published articles, whether for commercial or non-commercial purposes, provided that proper acknowledgment is given to the original author and the eProsiding Sistem Informasi (POTENSI) as the source of publication.
Any derivative works must be distributed under the same license. This license supports the open access principle to promote the free and responsible dissemination of scientific knowledge.
