Phishing Website Detection: A Comparative Analysis of Outlier Detection, Feature Selection, and Ensemble Algorithms

Öz

With the advancement of network technologies, phishing is a social engineering attack where attackers use fake websites to steal users' identities, and particularly their financial information. Machine learning methods have been widely used in recent years to detect these attacks. In this study, gradient-based ensemble models (GBM, XGB, LGBM, HGBM) were used for phishing detection. First, anomaly detection was performed on the original dataset using the iForest and IQR algorithms; then, sub-datasets named IFAS, IMIS, IQAS, and IQMIS were created using ANOVA and Mutual Information (MI) methods. Following hyperparameter optimization with Optuna, the XGB model achieved 95.67% and 92.87% accuracy on the IMIS and IQMIS datasets, respectively. The LGBM model stood out with a 95.53% accuracy rate on the IFAS dataset, while the highest accuracy on the IQAS dataset was achieved by the HGBM model at 93.66%. The results demonstrate that the XGB model has high potential for success in phishing detection.

Anahtar Kelimeler

• Ensemble Learning
• Optuna
• Phishing Detection
• Outlier Detection
• Gradient-Based Models

Etik Beyan

An ethical declaration was not required within the scope of the study.

Kimlik Avı Web Sitelerinin Tespiti: Aykırı Değer Tespiti, Özellik Seçimi ve Topluluk Öğrenme Algoritmalarının Karşılaştırmalı Analizi

Öz

Ağ teknolojilerinin gelişimiyle birlikte kimlik avı, saldırganların sahte web siteleri aracılığıyla kullanıcıların kimlik ve özellikle finansal bilgilerini ele geçirdiği bir sosyal mühendislik saldırısıdır. Bu saldırıların tespitinde son yıllarda makine öğrenmesi yöntemleri yaygın olarak kullanılmaktadır. Bu çalışmada, kimlik avı tespitinde gradyan tabanlı topluluk öğrenmesi modelleri (GBM, XGB, LGBM, HGBM) kullanılmıştır. İlk olarak, orijinal veri setinde iForest ve IQR algoritmalarıyla anomali tespiti yapılmış; ardından ANOVA ve Mutual Information (MI) yöntemleriyle IFAS, IMIS, IQAS ve IQMIS adlı alt veri setleri oluşturulmuştur. Optuna ile hiperparametre optimizasyonu sonrasında, XGB modeli IMIS ve IQMIS veri setlerinde sırasıyla %95,67 ve %92,87 doğruluk sağlamıştır. IFAS veri setinde %95,53 doğruluk oranı ile LGBM modeli öne çıkarken IQAS veri setinde ise en yüksek doğruluk %93,66 ile HGBM modelinde elde edilmiştir. Sonuçlar, XGB modelinin kimlik avı tespitinde yüksek başarı potansiyeline sahip olduğunu göstermektedir.

Anahtar Kelimeler

• Topluluk Öğrenmesi
• Optuna
• Kimlik Avı Tespiti
• Aykırı Değer Tespiti
• Gradyan Tabanlı Modeller

Etik Beyan

Çalışma kapsamında etik beyannameye ihtiyaç duyulmamıştır.

Kaynakça

1. Akiba T., Sano S., Yanase T., Ohta T., Koyama M. Optuna: A next-generation hyperparameter optimization framework. Proceedings of the 25th ACM SIGKDD International Conference on Knowledge Discovery and Data Mining 2019; 2623–2631.
2. Al Farizi WS., Hidayah I., Rizal MN. Isolation forest based anomaly detection: A systematic literature review. 8th International Conference on Information Technology, Computer and Electrical Engineering (ICITACEE) 2021; 118–122.
3. Ali G., Shah S., ElAffendi M. Enhancing cybersecurity incident response: AI-driven optimization for strengthened advanced persistent threat detection. Results in Engineering 2025; 25: 104078.
4. Aljofey A., Jiang Q., Rasool A., Chen H., Liu W., Qu Q., Wang Y. An effective detection approach for phishing websites using URL and HTML features. Scientific Reports 2022; 12(1): 8842.
5. Chen T., Guestrin C. XGBoost: A scalable tree boosting system. Proceedings of the 22nd ACM SIGKDD International Conference on Knowledge Discovery and Data Mining 2016; 785–794.
6. Chen Z., Zou H., Hu T., Yuan X., Fang X., Pan Y., Li J. HC-NIDS: Historical contextual information based network intrusion detection system in internet of things. Computers & Security 2025; 152: 104367.
7. Choudhary T., Mhapankar S., Bhddha R., Kharuk A., Patil R. Machine learning approach for phishing attack detection. Journal of Artificial Intelligence and Technology 2023; 3(3): 108-113.
8. Connolly A., Atlam HF. Effective ensemble learning phishing detection system using hybrid feature selection. Journal of Network and Computer Applications 2025; 242: 104251.

9. Dastjerdy B., Saeidi A., Heidarzadeh S. Review of applicable outlier detection methods to treat geomechanical data. Geotechnics 2023; 3(2): 375–396.
10. Dissanayake K., Md Johar MG. Comparative study on heart disease prediction using feature selection techniques on classification algorithms. Applied Computational Intelligence and Soft Computing 2021; 2021: 1–17.
11. Du W., Huang Q., Xu R. Follow the vine to get the melon: A deep framework for blockchain phishing fraud detection. Decision Support Systems 2025; 199: 114555.
12. Erendor ME., Yildirim M. Cybersecurity awareness in online education: A case study analysis. IEEE Access 2022; 10: 52319–52335.
13. Etem T., Teke, M. Advanced phishing detection: Leveraging t-SNE feature extraction and machine learning on a comprehensive URL dataset. Acta Infologica 2024; 8(2): 213–221.
14. Falch M., Olesen H., Skouby KE., Tadayoni R., Williams I. Cybersecurity strategies for SMEs in the nordic baltic region. Journal of Cyber Security and Mobility 2023; 11(6): 727-754.
15. Fang X., Xu M., Xu S., Zhao P. A deep learning framework for predicting cyber attacks rates. Eurasip Journal on Information Security 2019; 2019(1): 1–11.
16. Farnoudkia H. The inference of complicated networks by mutual information. Journal of Turkish Operations Management 2023; 7(1): 1591–1595.
17. Friedman JH. Greedy function approximation: A gradient boosting machine. The Annals of Statistics 2021; 29(5): 1189-1232.
18. GK Seda., Filiz A., GY., Sevda. AI-Driven predictions of mathematical problem-solving beliefs: Fuzzy logic, adaptive neuro-fuzzy inference systems, and artificial neural networks. Applied Sciences 2025; 15(2); 494.
19. Gürfidan R. Intelligent methods in cyber defence: Machine learning based phishing attacak detection on web pages. Mühendislik Bilimleri ve Tasarım Dergisi 2024; 12(2): 416–429.
20. Halbouni A., Gunawan TS., Habaebi MH., Halbouni M., Kartiwi M., Ahmad R. Mac hine learning and deep learning approaches for cybersecurity: A review. IEEE Access 2022; 10: 19572–19585.
21. Hassanali M., Soltanaghaei M., Taghi JG, Farsad ZB. Software development effort estimation using boosting algorithms and automatic tuning of hyperparameters with optuna. Journal of Software: Evolution and Process 2024; 36(9).
22. Hu L., Gao L., Li Y., Zhang P., Gao W. Feature-specific mutual information variation for multi-label feature selection. Information Sciences 2022; 593: 449–471.
23. Johnson KJ., Synovec RE. Pattern recognition of jet fuels: comprehensive GC×GC with ANOVA-based feature selection and principal component analysis. Chemometrics and Intelligent Laboratory Systems 2002; 60(1–2): 225–237.
24. K. Sruthi., Naik SM. A novel framework for effective phishing URL detection using an LSTM-based siamese network. Knowledge-Based Systems 2025; 329: 114271.
25. Ke G., Meng Q., Finley T., Wnag T., Chen W., Ma W., Ye Q., Liu, Y. LightGBM: A highly efficient gradient boosting decision tree. NIPS’17: Proceedings of the 31st International Conference on Neural Information Processing Systems 2017; 3149–3157.
26. Khanan A., Abdelgadir MY., Mohamed AHHM., Bashir M. From bytes to insights: A systematic literature review on unraveling IDS datasets for enhanced cybersecurity understanding. IEEE Access 2024; 12: 59289–59317.
27. Khoa TV., Hoang DT., Trung NL., Nguyen CT., Quynh TTT., Nguyen DN., Ha NV., Dutkiewicz E. Deep transfer learning: A novel collaborative learning model for cyberattack detection systems in IoT networks. IEEE Internet of Things Journal 2023; 10(10): 8578–8589.
28. Kilinç F., Eyüpoğlu C. Ağ ortamındaki saldırı türleri: Saldırı senaryo örnekleri. İstanbul Ticaret Üniversitesi Teknoloji ve Uygulamalı Bilimler Dergisi 2023; 6(1): 99–109.
29. Koukaras P., Mustapha A., Mystakidis A., Tjortjis C. Optimizing building short-term load forecasting: A comparative analysis of machine learning models. Energies 2024; 17(6): 1450.
30. Li D., Liu Z., Armaghani DJ., Xiao P., Zhou J. Novel ensemble intelligence methodologies for rockburst assessment in complex and variable environments. Scientific Reports 2022; 12(1): 1844.
31. Lin H., Ding H. Predicting ion channels and their types by the dipeptide mode of pseudo amino acid composition. Journal of Theoretical Biology 2011; 269(1): 64–69.
32. Mamdouh FH., Abd El-Hafeez T. A high-quality feature selection method based on frequent and correlated items for text classification. Soft Computing 2023; 27(16): 11259–11274.
33. Mandić-Rajčević S., Colosio C. Methods for the identification of outliers and their influence on exposure assessment in agricultural pesticide applicators: A proposed approach and validation using biological monitoring. Toxics 2019; 7(3): 37.
34. Mohd Ariffin NH., Mohamed Iqbal MI., Yusoff M., Mohd Zulkefli NA. A Study on the best classification method for an ıntelligent phishing website detection system. ASEAN Artificial Intelligence Journal 2025; 1(1): 20–33.
35. Mughaid A., AlZu’bi S., Hnaif A., Taamneh S., Alnajjar A., Elsoud EA. An intelligent cyber security phishing detection system using deep learning techniques. Cluster Computing 2022; 25(6): 3819–3828.
36. Nasiri H., Alavi SA. A novel framework based on deep learning and ANOVA feature selection method for diagnosis of COVID-19 cases from chest X-Ray images. Computational Intelligence and Neuroscience 2022; 2022: 1–11.
37. Ovı S. Phishing Website Detection Datasets. Kaggle 2025. Access Date: October 16, 2025. https://www.kaggle.com/datasets/mdsultanulislamovi/phishing-website-detection-datasets
38. Patra C., Giri D., Kundu B., Maitra T., Wazid M. Rhetorical structure theory-based machine intelligence-driven deceptive phishing attack detection scheme. Journal of Information Security and Applications 2025; 94: 104184.
39. Patro SP., Padhy N., Sah RD. An ensemble approach for prediction of cardiovascular disease using meta classifier boosting algorithms. International Journal of Data Warehousing and Mining 2022; 18(1): 1–29.
40. Postalcioglu S. Design of automatic tool for diagnosis of pneumonia using boosting techniques. Brazilian Archives of Biology and Technology 2022; 65: 1–16.
41. Saltık Ö. A sectoral perspective on frontrunners and information cascades: Central and isolated clearing analysis using isolation forest. Nişantaşı Üniversitesi Sosyal Bilimler Dergisi 2024; 12(Special Issue): 293–321.
42. Sánchez-Paniagua M., Fidalgo E., Alegre E., Alaiz-Rodríguez R. Phishing websites detection using a novel multipurpose dataset and web technologies features. Expert Systems with Applications 2022; 207: 118010.
43. Şenol A. Comparison of performance of classification algorithms using standard deviation-based feature selection in cyber attack datasets. International Journal of Pure and Applied Sciences 2023; 9(1): 209–222.
44. Shafin SS. An explainable feature selection framework for web phishing detection with machine learning. Data Science and Management 2025; 8(2): 127–136.
45. Shannon CE. A Mathematical theory of communication. Bell System Technical Journal 1948; 27(4): 623–656.
46. Singh SK., Jain R., Palaniappan D., Parmar K., Premavathi T., Gothania J. Spatiotemporal analysis and machine learning-based prediction of air quality in indian urban. Environmental Research and Technology 2024; 8(4): 809-822.
47. Taha A. Intelligent ensemble learning approach for phishing website detection based on weighted soft voting. Mathematics 2021; 9(21): 2799.
48. Toğaçar M. Web sitelerinde gerçekleştirilen oltalama saldırılarının yapay zekâ yaklaşımı ile tespiti. Bitlis Eren Üniversitesi Fen Bilimleri Dergisi 2021; 10(4): 1603–1614.
49. Tripathy N., Kumar BS., Parida S., Nayak SK. Cryptocurrency fraud detection through classification techniques. International Journal of Electrical and Computer Engineering (IJECE) 2024; 14(3): 2918.
50. Tuğrul B., Shihab A., Ahmed A. Makine öğrenme yöntemleri ile ağ trafik analizi. Nigde Omer Halisdemir University Journal of Engineering Sciences 2022; 11(4): 862–870.
51. Wilcox R. Estimating measures of location and scale. In Introduction to Robust Estimation and Hypothesis Testing (Third Edition) 2012; 43-101.