Perancangan Sistem Deteksi ARP Poisoning pada Wireless LAN Menggunakan Machine Learning: Random Forest dan AdaBoost

Ersamazaya, Rafi Dhafin (2025) Perancangan Sistem Deteksi ARP Poisoning pada Wireless LAN Menggunakan Machine Learning: Random Forest dan AdaBoost. Undergraduate thesis, UPN Veteran Jawa Timur.

	Text (Cover) Cover.pdf Download (14MB)
	Text (Bab 1) Bab 1.pdf Download (251kB)
	Text (Bab 2) Bab 2.pdf Restricted to Repository staff only until 2 December 2028. Download (623kB)
	Text (Bab 3) Bab 3.pdf Restricted to Repository staff only until 2 December 2028. Download (427kB)
	Text (Bab 4) Bab 4.pdf Restricted to Repository staff only until 2 December 2028. Download (4MB)
	Text (Bab 5) Bab 5.pdf Download (212kB)
	Text (Daftar Pustaka) Daftar Pustaka.pdf Download (250kB)
	Text (Lampiran) Lampiran.pdf Restricted to Repository staff only Download (811kB)

Abstract

ARP Poisoning attacks represent a serious threat to network security, wherein an attacker manipulates the Address Resolution Protocol (ARP) table with the intention of redirecting data traffic to an unauthorized device. Such attacks can cause significant impacts, including eavesdropping, data modification, and data theft within Wireless LAN environments. Therefore, an effective detection mechanism capable of accurately and efficiently identifying such activities is required to mitigate risks to network security. This thesis aims to design and evaluate an ARP Poisoning attack detection system using several machine learning algorithms, including a hybrid model combining Random Forest and AdaBoost, with the objective of comparing its effectiveness against four other models. The research employs the CRISP-DM (Cross-Industry Standard Process for Data Mining) methodology, which consists of the stages of business understanding, data understanding, data preparation, modeling, evaluation, and deployment. The dataset collection process resulted in 11,225 records composed of a combination of the Gelateria dataset and attack simulation data, which subsequently underwent EDA, cleaning, construction, formatting, and modeling. The modeling results indicate that the hybrid Random Forest–AdaBoost model achieved high accuracy, namely 99.92% on the validation data and 99.94% on the testing data, with a low error rate. This confirms the effectiveness of the hybrid model in detecting ARP Poisoning attacks. However, the hybrid model has a minimum inference time of 517.30 ms, which is relatively slow for detection tasks, making it less optimal for real-time implementation. Conversely, the AdaBoost model demonstrates more balanced performance, with an accuracy of 99.92% on the validation data and 99.94% on the testing data, and an error rate comparable to the hybrid model. This model exhibits a significantly faster inference time, with a maximum of 14.93 ms. Therefore, this thesis concludes that the AdaBoost model is the most suitable for real-time ARP Poisoning detection systems.

Item Type:

Thesis (Undergraduate)

Contributors:

Contribution	Contributors	NIDN/NIDK	Email
Thesis advisor	Arifiyanti, Amalia Anjani	NIDN0712089201	amalia_anjani.fik@upnjatim.ac.id
Thesis advisor	Kartika, Dhian Satria Yudha	NIDN0722058601	dhian.satria@upnjatim.ac.id

Subjects:

T Technology > T Technology (General)

Divisions:

Faculty of Computer Science > Departemen of Information Systems

Depositing User:

Rafi Dhafin Ersamazaya

Date Deposited:

03 Dec 2025 01:59

Last Modified:

03 Dec 2025 01:59

URI:

https://repository.upnjatim.ac.id/id/eprint/47457