Klasifikasi Deteksi Tautan Phishing Menggunakan Algoritma Machine Learning dan Metode Seleksi Fitur Berbasis Website

Classification of Phishing Link Detection Using Machine Learning Algorithms and Website-Based Feature Selection

Penelitian ini bertujuan untuk mengembangkan sistem klasifikasi deteksi tautan phishing dan legitimate menggunakan algoritma machine learning dan metode seleksi fitur berbasis website. Algoritma yang digunakan dalam penelitian ini adalah Random Forest dan Logistic Regression dengan metode seleksi fitur Chi-Square. Proses penelitian meliputi preprocessing data, ekstraksi fitur berbasis struktur URL dan konten website, seleksi fitur, hyperparameter tuning, serta evaluasi model klasifikasi. Kinerja model dievaluasi menggunakan beberapa metrik pengujian untuk menentukan model terbaik dalam mendeteksi tautan phishing. Hasil penelitian menunjukkan bahwa algoritma machine learning dan metode seleksi fitur mampu digunakan untuk mendeteksi tautan phishing secara efektif. Model terbaik kemudian diimplementasikan ke dalam sistem berbasis website sehingga dapat membantu pengguna mengidentifikasi tautan berbahaya secara realtime.

This research aims to develop a phishing and legitimate link detection classification system using machine learning algorithms and a websitebased feature selection method. The algorithms employed in this study are Random Forest and Logistic Regression, with Chi-Square used as the feature selection method. The research process includes data preprocessing, feature extraction based on URL structure and website content, feature selection, hyperparameter tuning, and classification model evaluation. Model performance is evaluated using several assessment metrics to determine the best-performing model for phishing link detection. The results indicate that machine learning algorithms combined with feature selection can effectively detect phishing links. The best-performing model is then implemented into a web-based system to assist users in identifying malicious links in real time.