Skripsi

PERBANDINGAN MACHINE LEARNING XGBOOST DAN DEEP LEARNING CNN 1D UNTUK MENDETEKSI PHISHING BERBASIS DATA URL

---

Phishing is a form of cyberattack that exploits fraudulent websites to obtain users’ sensitive information. This study aims to compare the performance of the XGBoost machine learning algorithm and CNN 1D deep learning in detecting phishing URLs using tabular URL data. XGBoost was chosen because it is effective in handling tabular data and capable of modeling nonlinear relationships among features, while CNN 1D was selected because it can automatically learn pattern representations through convolution. The dataset used in this study is StealthPhisher from Mendeley Data, consisting of 336,749 URLs with 65 features, including 160,943 legitimate URLs and 175,806 phishing URLs. The results show that the best configuration for CNN 1D was obtained by Model 3, with an accuracy of 0.998723, precision of 0.999033, recall of 0.998521, and F1-score of 0.998777. Meanwhile, the best configuration for XGBoost was also achieved by Model 3, with an accuracy of 0.999495, precision of 0.999431, recall of 0.999602, and F1-score of 0.999517. Based on these results, it can be concluded that XGBoost is the most superior model in this study for phishing detection using tabular URL data. Keywords: Phishing, XGBoost, CNN 1D, URL detection, machine learning, deep learning.

---

Availability

Detail Information

Series Title

-

Call Number

T1961092026

Publisher

Indralaya : Prodi Teknik Informatika, Fakultas Ilmu Komputer Universitas Sriwijaya., 2026

Collation

xvi, 61 hlm.; ilus.; 29 cm

Language

Indonesia

ISBN/ISSN

-

Classification

005.807

Content Type

Text

Media Type

-

Carrier Type

-

Edition

-

Subject(s)

Prodi Teknik Informatika
Phishing

Specific Detail Info

-

Statement of Responsibility

TUTI

Other version/related

File Attachment

Comments

---

You must be logged in to post a comment