Article Per Year (5 Year)
p-Index From 2021 - 2026
0.23
P-Index
This Author published in this journals
All Journal Jurnal Informatika dan Rekayasa Komputer
Chelsea Febiola br Sebayang
Universitas Negeri Medan
Computer Science & IT

Published : 1 Documents Claim Missing Document
Claim Missing Document
Check
Articles

Found 1 Documents
Search

 1 
Deteksi URL Phishing Menggunakan Hybrid Deterministic Finite Automaton dan Random Forest Yusrida Jelianti Sihite; Zulfahmi Indra; Chelsea Febiola br Sebayang; Jhon Gabriel Simarmata
Jurnal Informatika Dan Rekayasa Komputer(JAKAKOM) Vol 6 No 2 (2026): JAKAKOM Vol 6 No 2 SEPTEMBER 2026
Publisher : LPPM Universitas Dinamika Bangsa

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.33998/jakakom.2026.6.2.2901

Abstract

Phishing is one of the cybersecurity threats that exploits fake websites to obtain sensitive user information through URL manipulation resembling legitimate websites. Machine learning-based detection approaches have been widely used; however, most still suffer from limitations such as lack of classification transparency (black-box) and the potential for false positives on legitimate websites. This study aims to develop a phishing URL detection system using a Hybrid Deterministic Finite Automaton (DFA) and Random Forest approach to improve both detection accuracy and transparency. The study utilizes the PhiUSIIL dataset from the UCI Machine Learning Repository, consisting of phishing and legitimate URLs. The research stages include preprocessing, URL feature extraction, URL pattern analysis using DFA, classification using Random Forest, and model evaluation through Accuracy, Precision, Recall, F1-Score, and Confusion Matrix. The evaluation results indicate that the Hybrid DFA and Random Forest approach achieved the best performance with an Accuracy of 99.90%, Precision of 99.80%, Recall of 100.00%, and F1-Score of 99.90%. Furthermore, DFA integration improved system transparency through state transition visualization and significantly reduced both false positive and false negative classification errors compared to standalone models. Therefore, the proposed hybrid approach is effective for detecting phishing URLs more accurately and understandably for users.
Co-Authors Jhon Gabriel Simarmata Yusrida Jelianti Sihite Zulfahmi Indra, Zulfahmi