Intelligent Intrusion Detection System using Enhanced Arithmetic Optimization Algorithm with Deep Learning Model

doi:10.17559/tv-20221128071759

Cited by 7 publications

(4 citation statements)

References 19 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…This concept is an expansion of the one put forth in [37], which suggests a technique called CoDeepNEAT for creating optimized neural networks. An algorithmic approach for performing hyperparameter tuning is proposed in [50], where the arithmetic optimization algorithm is enhanced to improve the candidate solution's performance. The AutoML technique is used along with the ensembled strategy for the ideal selection of hyperparameters for the supervised setup using the voting mechanism proposed in [51].…”

Section: State Of the Art For Auto Parameter Tuningmentioning

confidence: 99%

Genetic Algorithm based hyperparameter tuned CNN for identifying IoT intrusions

2024

KSII TIIS

View full text Add to dashboard Cite

In recent years, the number of devices being connected to the internet has grown enormously, as has the intrusive behavior in the network. Thus, it is important for intrusion detection systems to report all intrusive behavior. Using deep learning and machine learning algorithms, intrusion detection systems are able to perform well in identifying attacks. However, the concern with these deep learning algorithms is their inability to identify a suitable network based on traffic volume, which requires manual changing of hyperparameters, which consumes a lot of time and effort. So, to address this, this paper offers a solution using the extended compact genetic algorithm for the automatic tuning of the hyperparameters. The novelty in this work comes in the form of modeling the problem of identifying attacks as a multi-objective optimization problem and the usage of linkage learning for solving the optimization problem. The solution is obtained using the feature map-based Convolutional Neural Network that gets encoded into genes, and using the extended compact genetic algorithm the model is optimized for the detection accuracy and latency. The CIC-IDS-2017 and 2018 datasets are used to verify the hypothesis, and the most recent analysis yielded a substantial F1 score of 99.23%. Response time, CPU, and memory consumption evaluations are done to demonstrate the suitability of this model in a fog environment.

show abstract

Section: State Of the Art For Auto Parameter Tuningmentioning

confidence: 99%

Genetic Algorithm based hyperparameter tuned CNN for identifying IoT intrusions

2024

KSII TIIS

View full text Add to dashboard Cite

show abstract

“…From time to time, it may be difficult for conventional intrusion detection systems to accurately identify variations in network traffic characteristics. Because of this, scientists emphasize how important it is to use ML and DL methods to increase IDS capabilities [1,[14][15][16]. A wide range of businesses, including finance, government, science, and security, are starting to apply ML and DL.…”

Section: Literature Reviewmentioning

confidence: 99%

Cloud Computing Security via Intelligent Intrusion Detection Mechanisms

Hamza Nasir,

Ayaz,

Nizamani

et al. 2024

ijisct

View full text Add to dashboard Cite

New challenges of data breaches, unauthorized access, and insider threats have emerged with the rise of Cloud Computing. Attackers find cloud systems attractive due to their common infrastructure. To combat these security threats, strong security procedures must be integrated. An intrusion detection system (IDS) is a vital instrument for safeguarding networks and cloud environments. It tracks system activity and network traffic. This study compares the efficacy of various machine learning models for intrusion detection. The Random-Forest-Classifier model performed the best with a test accuracy of 99.88% in recognizing known and unknown threats. Furthermore, various benefits have been identified when contrasting this intelligent intrusion detection system with conventional rule-based and signature-based methods. The use of machine learning in intrusion detection systems offers enhanced security measures and improved threat detection capabilities, highlighting the importance of leveraging advanced technologies to secure cloud environments. This research illuminates the necessity to employ cutting-edge techniques to protect cloud computing systems from malicious actions in the era of cybersecurity. More than that, it stresses the necessity to invest in enrichment mechanisms in machine learning-based intrusion detection systems to stay abreast with the evolving security problems.

show abstract

“…The Internet of Things (IoT) is mainly dependent upon interrelated smart devices, and various services can be employed to incorporate them as a single network [1]. It enables smart devices to collect confidential data and perform main functions, and these devices connect and interact with one another at maximum speed and make decisions based on indicator data [2].…”

Section: Introductionmentioning

confidence: 99%

Blockchain Assisted Fireworks Optimization with Machine Learning based Intrusion Detection System (IDS)

2024

Teh. vjesn.

View full text Add to dashboard Cite

In order to cope with the growing complexity of cyber attacks, it is imperative to have efficient intrusion detection systems (IDSs) that can monitor computer resources and produce data on abnormal or suspicious activities. The security of IoT networks is increasingly becoming a crucial concern as the Internet of Things (IoT) technology receives widespread use. Protecting the IoT framework with a conventional Intrusion Detection System (IDS) might be challenging due to the vast quantity and diversity of IoT devices. Traditional Intrusion Detection Systems (IDSs) face limitations when deployed in IoT networks due to resource limitations and the inherent complexity of these networks.This research proposed the Blockchain Assisted Fireworks Optimization with Machine Learning based Intrusion Detection System (BAFWO-MLIDS) technique in the healthcare platform. The major purpose of the BAFWO-MLIDS system is to apply BC technology (BCT) with IDS for enhanced security in the healthcare sector. The BCT enables to achieve secure data transmission in the healthcare platform. The BCT enables to achieve secure data broadcast in the healthcare environment. The BAFWO-MLIDS technique involves a three-stage procedure: FWO based FS process, ENN-based detection, and BO-based parameter optimization. In the proposed BAFWO-MLIDS technique, the FWO-based feature selection process is involved to select optimal features. For intrusion detection, the BAFWO-MLIDS technique uses Elman Neural Network (ENN) model. Finally, the Bayesian optimization (BO) technique is applied to modify the parameters compared with the ENN model and thereby it accomplishes enhanced detection performance. The simulation results of the BAFWO-MLIDS system can be inspected in a series of experiments and the obtained results ensured greater efficiency of the BAFWO-MLIDS methodology with other recent algorithms.

show abstract

Intelligent Intrusion Detection System using Enhanced Arithmetic Optimization Algorithm with Deep Learning Model

Cited by 7 publications

References 19 publications

Genetic Algorithm based hyperparameter tuned CNN for identifying IoT intrusions

Genetic Algorithm based hyperparameter tuned CNN for identifying IoT intrusions

Cloud Computing Security via Intelligent Intrusion Detection Mechanisms

Blockchain Assisted Fireworks Optimization with Machine Learning based Intrusion Detection System (IDS)

Contact Info

Product

Resources

About