K-Nearest Neighbor for Detecting DDoS Attacks in Smart Home IoT Networks - Dalam bentuk buku karya ilmiah

YUSUF IQBAL KUSUMA YUDA

Informasi Dasar

40 kali
25.04.6669
005.8
Karya Ilmiah - Skripsi (S1) - Reference

The extensive proliferation of IoT devices has led to a massive increase in IoT-based DDoS attacks, which is posing a critical threat to smart home networks and other IoT-connected ecosystems. Existing security solutions lack the ability to identify new and sophisticated IoT malware variants effectively. This paper presents a machine learning-based approach using the K-Nearest Neighbor (KNN) classifier and its Bagging ensemble variant to identify diverse DDoS attacks like SYN Flood, UDP Flood, and Mirai botnet variants in IoT network traffic. In this work, the CICIoT2023 dataset and correlation-based feature selection method are utilized to identify the most impactful attributes to categorize attacks. Experiments demonstrate that the baseline KNN model achieves outstanding accuracy, precision, recall, and F1-score greater than 97% with less computational overhead compared to the ensemble approach, thus being deployable in real-time in low-resource IoT environments. This article demonstrates the feasibility of lightweight machine learning models to realize effective and efficient intrusion detection in smart home IoT networks. Subsequent research will explore more complex feature engineering and ensemble learning frameworks for enhanced detection of novel attack behaviors.
Keyword : Distributed Denial of Service, IoT Botnets, Machine Learning, K-Nearest Neighbor, Smart Home Security
 

Subjek

COMPUTER NETWORK SECURITY
 

Katalog

K-Nearest Neighbor for Detecting DDoS Attacks in Smart Home IoT Networks - Dalam bentuk buku karya ilmiah
 
viii, 45p.: il,; pdf file
English

Sirkulasi

Rp. 0
Rp. 0
Tidak

Pengarang

YUSUF IQBAL KUSUMA YUDA
Perorangan
Vera Suryani
 

Penerbit

Universitas Telkom, S1 Informatika (International Class)
Bandung
2025

Koleksi

Kompetensi

 

Download / Flippingbook

 

Ulasan

Belum ada ulasan yang diberikan
anda harus sign-in untuk memberikan ulasan ke katalog ini