Jul 2026· International Journal of Engineering Research and Science & Technology· 0 citations
TL;DR
An intelligent cyberattack detection system that applies machine learning and deep learning techniques to classify network traffic as either normal or malicious, and demonstrates the potential of machine learningbased intrusion detection systems in improving network security and supporting the protection of modern smart environments.
Abstract
The rapid growth of the Internet of Things (IoT) has connected billions of smart devices, making daily life more convenient while also increasing the risk of cyber threats. Traditional security methods often struggle to identify complex and evolving attacks in largescale IoT environments. This project presents an intelligent cyberattack detection system that applies machine learning and deep learning techniques to classify network traffic as either normal or malicious. The collected IoT dataset is first pre-processed by handling missing values, converting categorical features into numerical form, and normalizing the data. Principal Component Analysis (PCA) is then used to reduce the number of features while preserving the most important information. Two classification models, namely a Multi-Layer Perceptron (MLP) neural network and a Random Forest classifier, are trained and evaluated using the processed dataset. Their performance is compared based on prediction accuracy and classification results. Experimental findings show that the neural network achieves superior accuracy and effectively detects various attack categories with minimal false predictions. The developed system offers a reliable, scalable, and efficient solution for securing IoT networks against cyber-attacks. This approach demonstrates the potential of machine learningbased intrusion detection systems in improving network security and supporting the protection of modern smart environments.
The Internet of Things (IoT) has changed the way modern devices connect. Billions of smart devices now work together in healthcare, transportation, factories, smart homes, and critical infrastructure. But IoT devices are small and often under-resourced. This makes them easy targets for serious attacks such as DDoS, botnets, spoofing, ransomware, and data breaches. Most intrusion detection systems (IDSs) still rely on signature or basic machine learning. These methods make it difficult to detect new or unknown attacks in rapidly changing IoT settings. The study develops a deep learning system that can detect cyber attacks early in IoT networks. The design includes a complete data processing stage, feature normalization, and a hybrid deep learning model. The model itself can find spatial and temporal patterns in network traffic. It uses bypass neural networks (CNNs) to extract features and learn from sequences using short-term long-term memory networks (LSTM). Together, they provide high recognition accuracy with low false alarm. The framework was tested on a public IoT penetration dataset. Accuracy, accuracy, memory, F1 score, and receiver undercrew operation (ROC-AUC) were verified. The results show that this method separates natural motion from different types of attacks with high accuracy. This makes it suitable for real-time use. Overall, deep learning intrusion detection provides a measurable, flexible, and effective way to protect IoT systems from ever-changing threats.
Omar Najeeb Ahmed, Ahmed Amer Mohammed· Libyan Journal of Applied an...· 0 citations
The fast development of internet of things (IoT) networks has led to an increased probability of cyberattacks. Intrusion detection systems (IDS) are needed for identifying unauthorised access and malicious activities in such dynamic environments. However, existing machine learning (ML) models failed to handle the complexity and variability of modern cyber threats. In this work, a hybrid deep learning (DL)-based anomaly detection model is presented for IoT cybersecurity. The model combines three types of features: (i) supervised feature extraction using linear discriminant analysis (LDA) to extract the most discriminative features, (ii) unsupervised feature learning through autoencoders to capture latent representations of the input data, and (iii) statistical features such as mean, variance, skewness, and kurtosis to learn input characteristics. The fused feature matrix is fed into a learning based echo state network (LBESN) for final detection. The parameters of the LBESN model are tuned using black eagle optimizer (BEO). Experimental results on standard intrusion detection datasets such as UNSW-NB15, KDD99, and InSDN show that the proposed model achieves superior performance in terms of accuracy, precision, recall, and F1-score compared to conventional DL techniques.
P. Palpandi, B. Sakthivel, M. Ponnrajakumari et al.· International Journal of Inf...· 0 citations
Traditional rule-based intrusion detection systems generally fail in identifying unknown or evolving threats; thus, automated and adaptive kinds of methods are crucial. Deep learning models provide promising solutions, but many recent studies depend on hybrid architecture, which increase the computational cost and reduce deploying ability on real-time or resource-limited systems. In this paper, we present and test a standalone LSTM model for multiclass cyberattack detection based on a CIC_IoT_Dataset2023, a recent labeled dataset that mirrors the actual network environment containing 33 attack categories. The dataset was extremely imbalanced as benign traffic accounted for most of the classes. To detect such attacks, we used the Synthetic Minority Oversampling Technique (SMOTE) to increase the frequency of less common types of address. The pre-processed dataset was then employed to train four models (RNN, CNN, DNN and the proposed LSTM) for performance analysis with sequential data. The proposed LSTM model achieved an accuracy between 2% and 7%. LSTM had good detection for frequent attacks and slow-changing patterns, which shows its capacity in learning long-lasting dependencies. The results demonstrate that a simple, lightweight standalone LSTM model can be used for effective and realistic intrusion detection without the need for complex hybrid architecture.
Jawad Hussain Awan, Misbah Safdar, Muhammad Ayaz Shirazi et al.· Italian National Conference...· 0 citations
The Internet of Things (IoT) technologies have been rapidly adopted in the field of Cyber-Physical Systems (CPS) and
have greatly enhanced the automation, connectivity and operational efficiency of industrial and critical infrastructure (ICI)
environments. The advent of greater device connectivity, however, has also grown the attack surface, leaving CPS environments
open to many different cyber threats. The traditional methods of intrusion detection are not effective in detecting more complex
and advanced attacks because the features are manually designed and cannot be easily adapted. In this publication, a Deep
Learning (DL)framework for security monitoring is presented in this study, which uses CNN to improve the intrusion detection
capability of the cyber-attack system in the IoT-enabled CPS environment. The intended framework accepts the network traffic
data out of the UNSW-NB15 dataset to learn complicated cyber security threats and identify attacks from normal data traffic. We
compared the proposed CNN model against a conventional Support Vector Machine (SVM) classifier. The experimental results
demonstrate that the CNN model surpasses the others across key metrics, including precision, accuracy, F1 score and recall. The
proposed approach has proved to be efficient in capturing hidden traffic characteristics and improves the reliability of cyber
threat detection framework in dynamic IoT-based CPS systems. The evaluation results confirm that deep learning methods can
be a scalable and efficient way to enhance the cybersecurity of next-generation cyber-physical infrastructures.
Sowjanya Samineni, D. Chiranjeevi· International Journal for Re...· 0 citations
The Internet of Things (IoT) has rapidly developed and, accordingly, opened the door for more cyber-attacks due to the increase in the number of connected devices and the amount of sensitive data transmitted over these networks. Classic Intrusion Detection Systems (IDS) are unable to recognize threats that are coming in newly created ways since they cannot analyze the different traffic patterns well. The proposed intelligent intrusion detection framework integrates Deep Learning (DL) with nature-inspired optimization for efficient attack classification. The Hybrid CNN–LSTM deep architecture is designed to extract and learn temporal and spatial features of the network traffic, while “Particle Swarm Optimization (PSO)” and “Ant Colony Optimization (ACO)” are used for optimal feature selection and dimensionality reduction on the UNSW-NB15 dataset. The results of this study show that the optimization process not only greatly improves the classification accuracy but also saves a lot of time in computations. The Hybrid CNN-LSTM model obtains a binary accuracy of 0.948, which is coupled with 0.948 for precision, recall, and F1-score. Additionally, it surpasses CNN and LSTM models used separately. In multiclass prediction, the hybrid method claims 0.720 accuracy, 0.510 precision, and 0.680 recall, thus showing a considerable improvement in the detection of varied IoT attacks.
Mayank Agarwal· Journal of Intelligent Decis...· 0 citations
The rapid growth of the Internet of Things (IoT) has significantly improved connectivity and automation across various domains. However, it has also increased the risk of cyber threats, particularly botnet attacks that compromise connected devices and disrupt network operations. Traditional machine learning techniques often struggle to detect sophisticated and evolving botnet attacks due to the complexity and high-dimensional nature of network traffic. This project proposes a Hybrid Deep Learning Model for Efficient IoT Botnet Attack Detection by integrating Artificial Neural Networks , Convolutional Neural Networks , Long Short-Term Memory, and Recurrent Neural Networks into a stacked ACLR framework. The proposed model leverages the feature extraction capability of CNN, the temporal learning strength of LSTM and RNN, and the classification power of ANN to accurately identify malicious network traffic. The model is trained and evaluated using the UNSW-NB15 dataset after appropriate preprocessing, including data cleaning, normalization, and label encoding. Experimental results demonstrate that the proposed approach achieves high detection accuracy, precision, recall, F1-score, and ROCAUC, outperforming several existing machine learning and deep learning methods. The proposed hybrid framework provides a robust, scalable, and reliable solution for real-time botnet attack detection, enhancing the security and resilience of modern IoT networks against evolving cyber threats.
MUNAGALA KUSUMA, Dr.M.Ramesh· International Journal of Eng...· 0 citations