Ensuring secure and efficient intrusion detection in Software-Defined Networking (SDN) within network virtualization is crucial for modern cybersecurity. In this context, this work presents an AI-powered hybrid deep learning model integrating CNN, LSTM, GRU, and a Transformer Encoder for feature selection. SMOTE is used to balance class distributions, therefore strengthening the model. With ROC-AUC values of 0.9628, and accuracy of 82%, therefore attesting to improved classification performance. For virtualized SDN settings, this method presents an adaptive intrusion detection, hence improving network security and dependability for useful cyber-defense purposes.

With the exponential growth of IoT networks, ensuring robust security has become increasingly critical, as these systems are vulnerable to various cyberattacks. Traditional methods often struggle to handle the massive data generated by IoT devices. This paper introduces an AI-driven, big data approach to intelligent attack detection for IoT networks. Utilizing the NSLKDD dataset, we employed PySpark for preprocessing and chi-square-based feature selection to identify the 15 most significant features, optimizing performance and reducing computational costs. The proposed model, based on XGBoost, achieved outstanding classification results with 98.93% accuracy, and precision, recall, and F1-score approaching 99 %. Comparative analysis against models like Random Forest and LightGBM confirmed its effectiveness, providing a scalable, accurate solution for IoT security.

Early detection of skin cancer is critical for effective treatment, and recent advancements in e-health systems have enabled the integration of machine learning models for accurate diagnosis. In this context, this work provides an optimal method employing MobileNet for feature extraction and Random Forest for classification to identify skin cancer in e-health systems. Hyperparameter tuning of the Random Forest model using Chaos Particle Swarm Optimization (C-PSO) helped to improve model performance. Comprising 2,357 photos from ISIC, the collection comprises many skin disorders including melanoma, basal cell carcinoma, and squamous cell carcinoma. With an AUC of 0.88, the suggested model outperformed approaches like Extra Trees, K-Nearest Neighbors, and Naive Bayes. Comparative findings on accuracy, F1-score, precision, and recall show how well our method improves skin cancer diagnosis for e-health uses.

Phishing attacks remain a critical threat to smart devices, especially in next-generation networks, where distributed environments pose unique challenges for detection. In this context, We propose an optimal federated learning system to solve this problem. We fine-tune hyperparameters at the server using the Fennec Fox Optimization (FFO) method, then distribute them to federated clients. Trained across ten rounds, the model obtained an accuracy of 94 %, with consistently exceeding 93 % in F1, precision, and recall. Comparative study of distributed and centralized losses revealed the robustness of the model, therefore strengthening our approach and enabling phishing detection in next-generation smart device networks on a scalable basis.

Recently, the use of smart IoT devices is increased. This increases the incidence of cyber attacks in IoT devices. In this context, this paper introduces an Attention-Enhanced Hybrid Model for the detection of attack in IoT. The proposed model integrates Success History Intelligent Optimizer (SHIO) for optimal feature selection and a CNN-LSTM-ASPP-based deep learning framework for attack classification. The proposed model used CNN to extract spatial features and LSTM to capture temporal dependencies. In addition of that, Atrous Spatial Pyramid Pooling (ASPP) for multi-scale feature refinement. Experimental evaluations achieve 88.3% accuracy and more then.90 of AUC curve for all the attack classes. The proposed model also outperforming GRU, LSTM, RNN, and Transformer models.

This paper introduces a Unet-based architecture for enhanced virtual try-on applications within the Metaverse, leveraging the rapid advancements in 6G technology. Our model, built on PyTorch 2.1.2 and tested on an NVIDIA Tesla P100-PCIE-16GB GPU, demonstrates remarkable proficiency in cloth detection, a critical aspect of virtual fitting rooms. We evaluate our model using a Kaggle dataset, achieving a significant accuracy of 96% and a Dice score above 1.65 in our tests, indicating a high degree of precision in garment segmentation. The synergy between our model’s deep learning capabilities and the high-speed, low-latency properties of 6G networks promises a revolutionary virtual try-on experience catering to the nuanced demands of digital fashion in the Metaverse ecosystem.

In response to the growing security demands of the 6G-enabled Metaverse, this paper introduces an efficient intrusion detection system utilizing a novel LSTM-GRU-based neural network. The model capitalizes on the sequential data processing prowess of LSTM and GRU layers to discern complex patterns indicative of cyber threats. Evaluated on a diverse dataset, the model architecture demonstrated significant accuracy improvements over traditional machine learning methods, achieving a 0.90 overall accuracy with high precision across various attack classes. Our results, presented through loss and accuracy trends, confusion matrices, and classification reports, attest to the model’s robustness and generalizability. This study proposes a promising approach to safeguarding the Metaverse-6G convergence, marking a step forward in predictive cybersecurity measures.

In this paper, we present a novel approach for wildlife animal detection in natural habitats using a deep learning model. Our model leverages a combination of Convolutional Neural Networks (CNNs) and Long Short-Term Memory (LSTM) networks, trained on aerial imagery captured by unmanned aircraft. We conducted a comprehensive 50-epoch training and testing regimen, and the results reveal the model's learning trajectory and performance. The findings demonstrate a progressive increase in testing accuracy, indicative of the model's improved ability to recognize and classify a diverse range of animal species. This research contributes to wildlife conservation and monitoring efforts, particularly in remote or inaccessible areas, by providing an automated and accurate means of animal detection. The model's proficiency in real-world scenarios is highlighted, making it a valuable tool for ecological research and conservation initiatives.

Currently, phishing attacks are posing great damage to the online community. As traditions, attack detection strategies are not effective against this new type of threat. Hence, there is a need for advanced attack detection techniques. In this context, this research proposed a hybrid deep learning and big data-based technique for phishing attack detection approach. Our proposed approach used Conv2d layers in sequence for analysis of the incoming traffic and predict its behavior. We used different parameters to measure our proposed approach. Through the use of the cuckoo optimization algorithm, the propsed approach achieves a high accuracy of 92%.

In the changing terrain of Internet of Things (IoT) security, especially in distributed systems, effective and fast virus detection is a challenging task. Using a Cascaded LSTM-GRU architecture, this work presents a Malware Detection Framework tuned for the special needs of edge and cloud computing. This method leverages Gated Recurrent Units (GRUs) for sequence data management and Long Short-Term Memory (LSTM) networks' strengths for temporal pattern recognition to increase the efficacy of malware detection. The empirical assessment of our methodology revealed remarkable classification accuracy and Fl scores. These findings demonstrate the framework's ability to greatly improve cybersecurity measures across smart computing systems, especially in edge and cloud computing environments, therefore marking major progress in the area of intelligent malware detection.

In the field of NetSoftIoT, where network softwarization converges with the proliferation of IoT devices, ensuring robust security in SDN environments is paramount. This paper presents a novel approach that integrates RandomForest for optimized feature selection and LSTM networks for attack detection. Our methodology capitalizes on the LSTM's sequential data processing capability to discern patterns indicative of DDoS attacks within network traffic with an accuracy of 83 %. Leveraging a dataset comprising varied traffic types, our model demonstrated precision in identifying DDoS traffic with a recall of 0.94. The results, validated by confusion matrices and classification reports, indicate the model's efficacy in maintaining network integrity against malicious threats. This study advances the frontier of cybersecurity in SDN, crucial for the burgeoning landscape of IoT applications.

Unmanned Aerial Vehicles (UAVs) have become an increasingly important part of modern military operations, surveillance, and disaster response. However, UAV networks are vulnerable to Distributed Denial of Service (DDoS) attacks, which can cause serious disruption to mission-critical operations. In this research, we propose a novel Machine Learning-based DDoS Mitigation Framework for UAV environment using Software-Defined Networks (SDN). The proposed framework utilizes SDN's programmability and centralized control capabilities to provide intelligent traffic management and filtering for UAV networks. Machine learning algorithms are used to analyze network traffic and detect DDoS attacks in real-time. Once an attack is detected, the framework can automatically steer traffic away from the affected network segments, isolate the affected devices, or block the malicious traffic altogether. We used KDDCup to train our machine-learning model. We also compare five machine-learning models (random forest, logistic regression, KNN, decision tree classifier, and XGBoost) to find the most accurate model. Our results show that the proposed framework can effectively mitigate DDoS attacks on UAV networks while maintaining low latency and minimal overhead. Overall, our research presents a novel approach to mitigating the threat of DDoS attacks on UAV networks using SDN and machine learning techniques. The proposed framework can help ensure UAVs' safe and reliable operation in mission-critical scenarios.

Cyber-Physical-Social System (CPSS) big data, the term that has been popularized over the past decade, is different from other types of big data because and is often used to refer to data sets that are too large or complex to be analyzed by traditional means. CPSS big data is specified as global historical and local real-time data. Due to the vast and heterogeneous nature of CPSS, this big data requires authentication control and security protocols. Blockchain technology has emerged as a promising solution for building secure and decentralized access control frameworks that facilitate data sharing and collaboration in CPSS. However, existing blockchain-based authentication control frameworks have scalability, privacy, and usability limitations. In this context, we proposed a secure authentication technique for CPSS that provide security to the system. Our proposed approach is lightweight and secure against different type of cyber attacks, such as replay attacks, and session hijacking attacks.

Phishing attacks remain a persistent and evolving cybersecurity threat, particularly in the context of 6G wireless networks. This paper introduces an innovative approach to combat phishing threats, leveraging advanced techniques tailored to the unique challenges of 6G environments. Our research focuses on enhancing the security posture of 6G networks by deploying adaptive defense mechanisms for real-time phishing attack detection and prevention. In this study, we employ cutting-edge deep learning models specifically customized to the 6G landscape. A multi-layer neural network architecture is utilized, fortified with advanced activation functions optimized for the dynamic nature of 6G wireless communication. The proposed model is trained on an extensive and diverse dataset, carefully curated to include phishing and legitimate activities specific to 6G networks, enabling robust learning and broad generalization.

This paper presents a novel deep learning-based approach to detect various cyber attacks within 6G wireless networks, encompassing DoS, probe attacks, and Sybil attacks. Leveraging the KDD Cup dataset and implementing our solution using PyTorch, our method demonstrates remarkable effectiveness, surpassing conventional techniques. Our results showcase the model’s adaptability to evolving attack patterns, underscoring its potential in bolstering the security of 6G wireless networks. This research significantly contributes to the field of intrusion detection in the 6G wireless networks landscape, offering insights into the application of deep learning to tackle emerging cyber threats. With the continuous advancement of 6G networks, our proposed approach stands as a pivotal means of safeguarding network integrity and availability against a spectrum of cyber attacks. This study not only furthers intrusion detection in 6G wireless networks but also highlights the pivotal role of deep learning in addressing the dynamic and evolving nature of cyber threats.

The danger of distributed denial of service (DDoS) attacks has grown in tandem with the proliferation of intelligent information systems. Because of the sheer volume of connected devices, constantly shifting network circumstances, and the need for instantaneous reaction, conventional DDoS detection methods are inadequate for the IoT. In this context, this study aims to survey the current state of the art in the topic by reading relevant articles found in the Scopus database, with a brief overview of the IoT and DDoS as this study examines neural networks and their applicability to DDoS detection. Finally, a decision tree-based model is developed for the detection of DDoS attacks. The analysis sheds light on the present trends and issues in this field and suggests avenues for further study.

This paper explores the potential of utilizing semantic web technologies to improve class management in Chinese schools. By analyzing a comprehensive dataset obtained from the Scopus database, the study investigates publication trends, document types, keyword distributions, and author contributions in the field of semantic web technologies for class management. The findings reveal a growing interest in this research area and highlight the benefits of semantic web technologies in personalized learning, information retrieval, collaboration, and assessment. The paper discusses the practical implications, challenges, and considerations for implementing semantic web technologies in Chinese schools. It aims to provide valuable insights for educators, researchers, policymakers, and educational technology practitioners interested in enhancing class management practices through the innovative use of semantic web technologies.

There is widespread agreement that information systems are the lifeblood of the global economy, providing businesses with a crucial competitive edge in international markets and giving the world's governments and corporations of the 21st century the foundation they need to ensure the safety of their citizens and deliver superior services. Despite their importance, information systems face significant risks within and outside of the company. The best security methods can be circumvented by professional hackers, as the complexity of security attacks, the number of vulnerabilities, and the absence of implementation of quality information security management within companies have expanded substantially in recent years. For the sake of this discussion, the authors examine several security paradigms designed by scientists to protect the operations of multinational corporations.

About 70% of the earth's surface is covered with water, and providing safe transportation in this vast area is the responsibility of the maritime transportation system. The maritime transportation system provides safety to the vessel's passengers, manages the route of the vessels to avoid collision, and monitors the vessel's condition during the journey. All tasks performed by the maritime transport system require continuous monitoring of the vessel performance, and for this, different IoT devices are installed on the vessel board. The IoT devices collect data related to the physical parameters of the vessel and share it with the coast-side servers. As the shared data contains confidential information about the vessel, there is a need for secure data sharing techniques, allowing only the authenticated personals to access the data received from the maritime IoT devices. In this context, we developed an identity-based secure information sharing scheme for the maritime transport system that maintains proper authentication measures. Our proposed approach uses the concept of identity-based encryption for authentication management. Our proposed approach is IND-sID-CCA secure and works efficiently with maritime IoT devices.

Intelligent Transport Systems (ITS) is a developing technology that will significantly alter the driving experience. In such systems, smart vehicles and Road-Side Units (RSUs) communicate through the VANET. Safety apps use these data to identify and prevent hazardous situations in real-time. Detection of malicious nodes and attack traffic in Intelligent Transportation Systems (ITS) is a current research subject. Recently, researchers are proposing graph-based machine learning techniques to identify malicious users in the ITS environment, through which it is easy to analyze the network traffic and detect the malicious devices. Therefore, graph-based machine learning techniques could be a technique that efficiently detect malicious nodes in the ITS environment. In this context, this article aims to provide a technique for resolving authentication and security issues in ITS using lightweight cryptography and graph-based machine learning. Our solution uses the concepts of identity based authentication technique and graph-based machine learning in order to provide authentication and security to the smart vehicle in ITS. By authenticating smart vehicles in ITS and identifying various cyber threats, our proposed method substantially contributes to the development of intelligent transportation communication environment.

The purpose of this study is to provide an overview of distributed denial of service (DDoS) attack detection in intelligent systems. In recent times, due to the endemic COVID-19, the use of intelligent systems has increased. However, these systems are easily affected by DDoS attacks. A DDoS attack is a reliable tool for cyber-attackers because there is no efficient method which can detect or filter it properly. In this context, we analyze different types of DDoS attacks and defense techniques for intelligent systems. For the analysis, we used Scopus databases to collect relevant papers in English between 2014 and 2022. This study makes an important contribution to the field of DDoS attack detection for intelligent systems, providing a comprehensive overview of the field's evolution and current status, as well as a comprehensive, synthesized, and organized summary of various perspectives, definitions, and trends in the field.

Educators have been calling for reform for a decade. Recent technical breakthroughs have led to various improvements in the semantic web-based education system. After last year's COVID-19 outbreak, development quickened. Many countries and educational systems now concentrate on providing students with online education, which differs greatly from traditional classroom education. Online education allows students to learn at their own pace and the system. As a consequence, we may say that education has become more dynamic. In the educational system, this changing nature makes user demands difficult to identify. Many instructors suggest using machine learning, artificial intelligence, or ontology to improve traditional teaching methods. Due to the lack of survey studies examining and comparing all of the researcher's semantic web-based teaching methodologies, we decided to conduct this survey. This paper's goal is to analyse all available possibilities for semantic web-based education systems that enable new researchers to develop their knowledge.

The smart ocean has aroused the interest of government, business, and academia because of the wealth of marine resources. It has been suggested to use underwater Internet of Things (IoT) frameworks to collect a variety of data from smart seas that can aid in the underwater green transport system, ecological sustainability, military intelligence gathering, and a variety of other operations. Because of the limited resources accessible to IoT devices regarding communication overhead, processing expenses, and battery capacity, security and privacy concerns in underwater green transport systems have lately been a critical source of worry. In this context, We presented a unique identity-based authentication mechanism for underwater green transport systems. Our suggested solution uses lightweight authentication mechanisms that prove secure communication between different elements of the green transport system.

The area of intelligent transport systems (ITS) is attracting growing attention because of the integration of the smart IoT with vehicles that improve user safety and overall travel experience. Vehicular ad hoc network (VANET) is the part of ITS; that deals with the routing protocols and security of smart vehicles. However, due to the rapid increase in the number of smart vehicles, the existing network technology’s resources unable to handle the traffic load. It expects that the 6G communication system has the ability to fulfill the requirements of VANETs. Only a few studies explore this area, but they also overlooked the security aspect of VANETs in 6G communications networks. In this paper, we present an approach to address authentication and security issues for vehicles in VANET. By authenticating cars in the VANET and identifying various cyber assaults such as DDoS, our method significantly contributes to the intelligent transport communication network. Our approach uses the concepts of identity-based encryption to provide access control to the vehicles and deep learning-based techniques for filtering malicious packets. Our identity-based encryption technique is IND-sID-CCA secure, and a state-of-the-art deep learning algorithm detects malicious packets with an accuracy of 99.72%. These results emphasize the validity of our proposed approach for VANETs in 6G communication systems.