Implementasi Algoritma Random Forest untuk Deteksi Phishing pada Email

Abstrak

Email phishing continues to pose a significant cyber threat that compromises user data and financial security. This research proposes the implementation of the Random Forest algorithm to automatically classify phishing and legitimate emails. The dataset, consisting of 39,157 email records, was collected from Kaggle and independent email samples. The research includes data preprocessing, feature extraction using TF-IDF, model training with Random Forest, and system implementation in a web-based interface. Experimental results demonstrate that the model achieves 99.61% accuracy on the Kaggle dataset and 94% accuracy on independent test data. The system is designed to help users identify malicious emails efficiently through an intuitive web application. Keywords — Phishing, Random Forest, Machine Learning, Email Security

Identifier

• DOI: https://doi.org/10.56622/sentineljournal.v6i2.84
• OpenAlex: https://openalex.org/W7172342861