DEIGASe: Deep Extraction and Information Gain for an Optimal Anomaly Detection in IoT-based Smart Cities

Hazman, Chaimae; Guezzaz, Azidine; Benkirane, Said; Azrour, Mourade

doi:10.21203/rs.3.rs-2141835/v1

Cited by 6 publications

(5 citation statements)

References 28 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…This study [12] proposes the DEIGASe model, which combines deep extraction with feature selection to handle anomaly detection problems in IoT-based smart cities. Using a stacked auto-encoder, deep extraction is utilized to extract features from the IoT data.…”

Section: Related Workmentioning

confidence: 99%

See 1 more Smart Citation

SmartSentry: Cyber Threat Intelligence in Industrial IoT

Sadhwani,

Modi,

Muthalagu

et al. 2024

IEEE Access

View full text Add to dashboard Cite

While the Internet of Things (IoT) paradigm has transformed connectivity, it has also brought with it previously unheard-of security risks. The categorization of IoT attacks using several machine learning techniques and a deep learning method is the main emphasis of this research. In addition to proposing a binary and multiclass classification framework with Machine Learning (ML) algorithms like Random Forest (RF), Decision tree (DT), Extra Tree Classifier (ETC), Support Vector Machine (SVM), and k-Nearest Neighbor (KNN) and Deep Learning (DL) architectures like Deep Neural Network (DNN), the study assesses a wide range of attack types in IoT environments. Benchmark datasets with real-world IoT attack scenarios, such as Edge-IIoTset, are used for experimentation. Preprocessing is done on the dataset using Principal Componenet Analysis (PCA) for feature selection, Synthetic Minority Oversampling Technique to handle class imbalance and Standard Scaling for feature scaling. These approaches' comparative performance and efficacy are examined. The outcomes indicate how successful the DL model in managing intricate attack patterns and the generalization capabilities of ML algorithms across various attack classes. The DNN model yields the best results, with 100% accuracy for binary classification, 96.15% accuracy for 6-class classification, and 94.68% accuracy for 15-class classification. Further, 10-fold cross validation has been applied to make sure that the model does not overfit. This work contributes to the improvement of IoT security mechanisms by offering insights into the selection of appropriate approaches for binary and multiclass classification of threats.

show abstract

Section: Related Workmentioning

confidence: 99%

“…In[12], a deep extraction approach has been combined with feature selection. Although, the results suggest zero FPR, work has not been done on 6-class or 15-class classification.…”

mentioning

confidence: 99%

SmartSentry: Cyber Threat Intelligence in Industrial IoT

Sadhwani,

Modi,

Muthalagu

et al. 2024

IEEE Access

View full text Add to dashboard Cite

show abstract

“…Furthermore, meta-learning can help overcome the scalability challenges of ensemble-based IDSs in IoMT and IoT networks by optimizing computational resources and improving efficiency in handling large datasets and real-time applications [ 29 , 30 ]. By incorporating meta-learning techniques, IDSs can enhance their generalization capabilities across diverse datasets and domains, ensuring consistent performance and adaptability to varying network conditions and security threats [ 31 , 32 ]. Additionally, meta-learning can contribute to the development of robust and efficient security measures in the IoMT, addressing privacy concerns, data integrity issues, and the interoperability of security mechanisms across connected devices [ 33 , 34 ].…”

Section: Introductionmentioning

confidence: 99%

Enhancing the Internet of Medical Things (IoMT) Security with Meta-Learning: A Performance-Driven Approach for Ensemble Intrusion Detection Systems

Alalhareth,

Hong

2024

Sensors

View full text Add to dashboard Cite

This paper investigates the application of ensemble learning techniques, specifically meta-learning, in intrusion detection systems (IDS) for the Internet of Medical Things (IoMT). It underscores the existing challenges posed by the heterogeneous and dynamic nature of IoMT environments, which necessitate adaptive, robust security solutions. By harnessing meta-learning alongside various ensemble strategies such as stacking and bagging, the paper aims to refine IDS mechanisms to effectively counter evolving cyber threats. The study proposes a performance-driven weighted meta-learning technique for dynamic assignment of voting weights to classifiers based on accuracy, loss, and confidence levels. This approach significantly enhances the intrusion detection capabilities for the IoMT by dynamically optimizing ensemble IDS models. Extensive experiments demonstrate the proposed model’s superior performance in terms of accuracy, detection rate, F1 score, and false positive rate compared to existing models, particularly when analyzing various sizes of input features. The findings highlight the potential of integrating meta-learning in ensemble-based IDS to enhance the security and integrity of IoMT networks, suggesting avenues for future research to further advance IDS performance in protecting sensitive medical data and IoT infrastructures.

show abstract

“…Deep learning is a creative ML technique that improves and influences artificial neural network systems (ANNs). An ensemble‐based intrusion detection technique that integrates features along with genetic algorithms (GA), and information gain (IG) in favor of feature selection (DEIGASe) that determines the Xgboost, K‐NN, and MLP 7 . The stealing of personal issues has been mitigated by Wi‐Fi‐protected access (WPA) and wire equivalent protection (WEP) protocols 8 .…”

Section: Introductionmentioning

confidence: 99%

“…An ensemble-based intrusion detection technique that integrates features along with genetic algorithms (GA), and information gain (IG) in favor of feature selection (DEIGASe) that determines the Xgboost, K-NN, and MLP. 7 The stealing of personal issues has been mitigated by Wi-Fi-protected access (WPA) and wire equivalent protection (WEP) protocols. 8 The design of four layered convolutional neural networks regarding the time domain for the detection of replay attacks.…”

mentioning

confidence: 99%

A novel‐cascaded ANFIS‐based deep reinforcement learning for the detection of attack in cloud IoT‐based smart city applications

Almasri

Alajlan

2023

Concurrency and Computation

View full text Add to dashboard Cite

SummaryVast usages of Internet of Things (IoT) devices in various smart applications have laid a foundation for the evolution of modern smart cities. The increasing dependency of smart city applications on communication and information technologies enhances operational efficiency, sustainability, and automation of city services. However, due to the heterogeneous nature of IoT devices, the network faces critical security issues while executing continued network operations and services, particularly by cyber‐attacks. One of the predominant and rampant cyber‐attacks in smart city applications is botnet attacks. Therefore, a novel deep learning model for the detection and isolation of cyber‐attacks is proposed in the cloud IoT‐based smart city applications to protect against such cyber‐attacks. The proposed framework utilizes two different modules to automatically detect and isolate the malicious traffic emanating from compromised IoT devices with more efficiency. Here, two different datasets namely the IoT network intrusion and the ISCX 2012 IDs datasets are utilized for the evaluation of the proposed framework. In the first phase, the compromised device which communicates malicious network traffics through the network is identified using a cascaded adaptive neuro‐fuzzy inference system (CANFIS). After detection, IP address of abnormal traffic is recorded and informed to the system administrator. In the second phase, communication pathways of compromised devices with other normal devices are blocked and the compromised devices are isolated from the network using the modified deep reinforcement learning (MDRL) approach. The analytic result shows that the proposed framework achieves a greater accuracy rate of about 98.7% as compared to other state‐of‐art methods.

show abstract

DEIGASe: Deep Extraction and Information Gain for an Optimal Anomaly Detection in IoT-based Smart Cities

Cited by 6 publications

References 28 publications

SmartSentry: Cyber Threat Intelligence in Industrial IoT

SmartSentry: Cyber Threat Intelligence in Industrial IoT

Enhancing the Internet of Medical Things (IoMT) Security with Meta-Learning: A Performance-Driven Approach for Ensemble Intrusion Detection Systems

A novel‐cascaded ANFIS‐based deep reinforcement learning for the detection of attack in cloud IoT‐based smart city applications

Contact Info

Product

Resources

About