Makine ve derin öğrenme yaklaşımlarına dayalı saldırı tespit sistemi
Küçük Resim Yok
Tarih
2024
Yazarlar
Dergi Başlığı
Dergi ISSN
Cilt Başlığı
Yayıncı
Düzce Üniversitesi
Erişim Hakkı
info:eu-repo/semantics/openAccess
Özet
Bu tez çalışmasında, nesnelerin interneti (Internet of Things - IoT) ağlarına yönelik olarak PBCSC-IoT, PMBL-IoT ve AOCBiT-IoT isimli üç farklı saldırı tespit sistemi (Intrusion Detection System - IDS) geliştirilmiş ve bu modellerin başarımları kapsamlı bir şekilde analiz edilerek karşılaştırılmıştır. Önerilen IDS modellerinde kullanılan öznitelik indirgeme, veri dengeleme ve sınıflandırma yöntemlerinin IDS'ler üzerindeki etkileri değerlendirilmiş ve bu modellerin saldırı tespit başarımları karşılaştırılmıştır. PBCSC-IoT modeli, yarasa optimizasyon algoritması (Bat Optimization Algorithm -BAT), temel bileşenler analizi (Principal Component Analysis - PCA), sentetik azınlık aşırı örnekleme tekniği (Synthetic Minority Oversampling Technique - SMOTE) ve evrişimli sinir ağları (Convolutional Neural Network - CNN) yöntemlerinin hibrit bir yaklaşımla birleştirilmesiyle oluşturulmuş ve IoTID20 ve BoT-IoT verisetlerinde test edilmiştir. PMBL-IoT modeli; PCA, yusufçuk optimizasyon algoritması (Mayfly Optimization Algorithm - MAO), sınırda-SMOTE (Borderline SMOTE - BSMOTE) ve uzun kısa süreli bellek (Long Short-Term Memory - LSTM) yöntemleri ile geliştirilmiş olup IoTID20, CIC-ToN-IoT ve USB-IDS-1 verisetlerinin birleştirilmesiyle elde edilen birleşik veriseti üzerinde test edilmiştir. AOCBiT-IoT ise, Otokodlayıcı, Varyans analizi F-testi (Analysis of Variance F-Test - ANOVA F-Test), uyarlanabilir sentetik örnekleme yöntemi (Adaptive Synthetic Sampling - ADASYN), tek taraflı seçim (One-Sided Selection - OSS) yöntemi ile birlikte CNN, çift yönlü LSTM (Bidirectional LSTM - BiLSTM) ve Transformer yöntemleri ile geliştirilmiş olup, CIC-DDoS2019 veriseti üzerinde test edilmiştir. Sonuçlar, hibrit DL tekniklerinin IoT ağlarında saldırı tespitinde yüksek başarım sağladığını ve veri dengeleme ile öznitelik indirgeme yöntemlerinin bu sınıflandırma başarımını daha da artırabileceğini göstermiştir.
In this thesis, three different intrusion detection systems (IDS) named PBCSC-IoT, PMBL-IoT, and AOCBiT were developed for Internet of Things (IoT) networks, and their performances were thoroughly analyzed and compared. The effects of the feature reduction, data balancing, and classification methods used in the proposed IDS models on IDS performance were evaluated, and the detection success of these models was compared in detail. The PBCSC-IoT model was created using a hybrid approach that combines the Bat Optimization Algorithm (BAT), Principal Component Analysis (PCA), Synthetic Minority Oversampling Technique (SMOTE), and Convolutional Neural Network (CNN) methods, and was tested on the IoTID20 and BoT-IoT datasets. The PMBL-IoT model was developed using PCA, Mayfly Optimization Algorithm (MAO), Borderline SMOTE (BSMOTE), and Long Short-Term Memory (LSTM) methods and was tested on a combined dataset obtained by merging the IoTID20, CIC-ToN-IoT, and USB-IDS-1 datasets. The AOCBiT-IoT model was developed using an Autoencoder, Analysis of Variance F-Test (ANOVA F-Test), Adaptive Synthetic Sampling (ADASYN), One-Sided Selection (OSS) method along with CNN, Bidirectional LSTM (BiLSTM), and Transformer methods, and was tested on the CIC-DDoS2019 dataset. The results demonstrated that hybrid deep learning techniques provide high performance in detecting attacks in IoT networks and that data balancing and feature reduction methods can further enhance this classification performance.
In this thesis, three different intrusion detection systems (IDS) named PBCSC-IoT, PMBL-IoT, and AOCBiT were developed for Internet of Things (IoT) networks, and their performances were thoroughly analyzed and compared. The effects of the feature reduction, data balancing, and classification methods used in the proposed IDS models on IDS performance were evaluated, and the detection success of these models was compared in detail. The PBCSC-IoT model was created using a hybrid approach that combines the Bat Optimization Algorithm (BAT), Principal Component Analysis (PCA), Synthetic Minority Oversampling Technique (SMOTE), and Convolutional Neural Network (CNN) methods, and was tested on the IoTID20 and BoT-IoT datasets. The PMBL-IoT model was developed using PCA, Mayfly Optimization Algorithm (MAO), Borderline SMOTE (BSMOTE), and Long Short-Term Memory (LSTM) methods and was tested on a combined dataset obtained by merging the IoTID20, CIC-ToN-IoT, and USB-IDS-1 datasets. The AOCBiT-IoT model was developed using an Autoencoder, Analysis of Variance F-Test (ANOVA F-Test), Adaptive Synthetic Sampling (ADASYN), One-Sided Selection (OSS) method along with CNN, Bidirectional LSTM (BiLSTM), and Transformer methods, and was tested on the CIC-DDoS2019 dataset. The results demonstrated that hybrid deep learning techniques provide high performance in detecting attacks in IoT networks and that data balancing and feature reduction methods can further enhance this classification performance.
Açıklama
Anahtar Kelimeler
Bilgisayar Mühendisliği Bilimleri-Bilgisayar ve Kontrol, Computer Engineering and Computer Science and Control
