Most existing research on decentralized IoT applications tends to address specific vulnerabilities, with relatively few techniques dedicated to managing privacy and trust issues. To mitigate these challenges, blockchain-based solutions are increasingly adopted to enhance the reliability and security of IoT networks. In particular, blockchain-based authentication frameworks offer a decentralized approach to storing and verifying device identities, enabling secure and trustless communication among devices and with external systems. However, current blockchain-based IoT systems often suffer from complexity and storage overhead. To address these limitations, we propose a novel solution tailored for large-scale IoT environments using a permissioned blockchain. Our approach incorporates optimized data storage and a lightweight authentication mechanism, offering improved scalability and reduced storage demands. Furthermore, we introduce, for the first time, the integration of homomorphic encryption to secure IoT data at the user end before uploading it to the cloud. The proposed framework is evaluated through comprehensive simulations and compared with existing benchmark models. This research contributes a trust-aware security model that significantly enhances both the privacy and security of IoT services.

The introduction of mobile banking is essential in today’s financial sector, where technological innovation plays a critical role. To remain competitive in the current market, businesses must analyze client attitudes and perspectives, as these influence long-term demand and overall profitability. While previous studies have explored general adoption behaviors, limited research has examined how individual factors such as social influence, lifestyle compatibility, financial technology self-efficacy, and perceived usage cost affect mobile banking adoption among specific generational cohorts. This study addresses that gap by offering insights into these variables, contributing to the growing literature on mobile banking adoption, and presenting actionable recommendations for financial institutions targeting younger market segments. Using a structured questionnaire survey, data were collected from both users and non-users of mobile banking among the Gen Z population in the United States. The regression model significantly predicts mobile banking adoption, with an intercept of $-0.548$ ($p < 0.001$). Among the independent variables, perceived cost of usage has the strongest positive effect on adoption ($B=0.857$, $\beta =0.722$, $p < 0.001$), suggesting that adoption increases when mobile banking is perceived as more affordable. Social influence also has a significant positive impact ($B=0.642$, $\beta =0.643$, $p < 0.001$), indicating that peer influence is a central driver of adoption decisions. However, self-efficacy shows a significant negative relationship ($B=-0.343$, $\beta =-0.339$, $p < 0.001$), and lifestyle compatibility was found to be statistically insignificant ($p=0.615$). These findings suggest that reducing perceived costs, through lower fees, data bundling, or clearer communication about affordability, can directly enhance adoption among Gen Z consumers. Furthermore, leveraging peer influence via referral rewards, Partnerships with influencers, and in-app social features can increase user adoption. Since digital self-efficacy presents a barrier for some, banks should prioritize simplifying user interfaces and offering guided assistance, such as tutorials or chat-based support. Future research may employ longitudinal designs or analyze real-life transaction data for a more objective understanding of behavior. Additional variables like trust, perceived risk, and regulatory policies, not included in this study, should be integrated into future models to offer a more comprehensive analysis.

The main objective of this study is to conduct a comprehensive analysis of cyber risks in cloud computing, including classifying threats, vulnerabilities, impacts, and countermeasures. This classification helps to identify suitable security controls to mitigate cyber risks for each type of threat. Additionally, this study aims to explore the main vulnerabilities in terms of infrastructure, service and platform in cloud computing. This study uses the content analysis technique to collect, analyze, and classify data in terms of types of threats, vulnerabilities, and countermeasures. The methodology comprises four primary stages: (1) identifying key components, (2) threat identification, (3) vulnerability identification, and (4) countermeasure identification. The results indicate that DoS attacks and account hijacking attacks were the most prevalent infrastructure vulnerabilities in cloud computing, each accounting for 14% and 10% of incidents. The results found that unpatched software and weak access controls were classified as the most critical threats in the service level in cloud computing, comprising 17% and 12% of incidents, respectively. The results also indicated that encryption methods, access controls mechanisms and firewall malware protection are the most significant and effective countermeasures for protecting the infrastructure, service and platform in cloud computing environment. The findings of this study provides valuable recommendations for academic research in classifying the different types of cyber threats and understanding the significant security controls against cyber-attacks in cloud computing.

The rapid expansion of the Internet of Things (IoT) has introduced new vulnerabilities that traditional security mechanisms often fail to address effectively. Signature-based intrusion detection systems cannot adapt to zero-day attacks, while rule-based solutions lack scalability for the diverse and high-volume traffic in IoT environments. To strengthen the security framework for IoT, this paper proposes a deep learning-based anomaly detection approach that integrates Convolutional Neural Networks (CNNs) and Bidirectional Gated Recurrent Units (BiGRUs). The model is further optimized using the Moth–Flame Optimization (MFO) algorithm for automated hyperparameter tuning. To mitigate class imbalance in benchmark datasets, we employ Generative Adversarial Networks (GANs) for synthetic sample generation alongside Z-score normalization. The proposed CNN–BiGRU + MFO framework is evaluated on two widely used datasets, UNSW-NB15 and UCI SECOM. Experimental results demonstrate superior performance compared to several baseline deep learning models, achieving improvements across accuracy, precision, recall, F1-score, and ROC–AUC. These findings highlight the potential of combining hybrid deep learning architectures with evolutionary optimization for effective and generalizable intrusion detection in IoT systems.

The emergence of the fourth industrial revolution has resulted in a fundamental change in the way organisations function, highlighting the need to make decisions based on data. Automated Machine Learning (AutoML) has become a key tool in this revolution, allowing quick model building and deployment without data science skills. This article examines how AutoML simplifying data-driven industrial application development affects efficiency, precision, and scalability. AutoML basics and industrial applications. AutoML tools automate data preparation, feature selection, model selection, hyperparameter tweaking, and model assessment for machine learning. Automation saves time and effort in developing effective ML models. Compared to more conventional white-box modelling methods, machine learning has often been shown to be the better choice in practical industrial settings. Still, development-related human variables reduce the predictability of locating a solution near the theoretical optimum. In contrast, Hyperparameter optimisation, model selection, and feature extraction may all be fully automated using automated machine learning (Auto ML). The article shows how AutoML can simplify the creation of data-driven industrial apps using a well-known open dataset. AutoML is a breakthrough tool that can expedite data-driven industrial application development. AutoML democratises sophisticated analytics and helps enterprises maximise data to use for smarter, more effective operations by automating difficult machine learning procedures.

The high rate of social media development has triggered a high rate of fake accounts, which are a great risk to the privacy of users and the integrity of the platform. These malicious accounts are hard to detect because user activity data is highly imbalanced, dimensional, and sequential. The emergence of fake profiles on social media endangers the privacy and trust of social media users. It is difficult to detect such accounts because of high-dimensional, highly sequential, and imbalanced user behavior data. Current techniques tend to miss out on the complicated activity patterns or even overfit, which is why a strong, scalable, and precise model of social media fraud detection is required. This study suggests a new deep learning architecture that entails a Temporal Convolutional Network (TCN) with Generative Adversarial Network (GAN)-based data augmentation to generate minority classes, and Autoencoder-based feature extraction to reduce dimensionality. The Seagull Optimization Algorithm (SOA), which is a metaheuristic algorithm, is used to optimize hyperparameters by balancing efficiency and speed of convergence in global search. The framework is tested on benchmark datasets (Cresci-2017 and TwiBot-22) and compared to the state-of-the-art models. It has been shown in experiments that the suggested TCN-GAN-SOA framework performs better, with ROC-AUC scores of 0.96 on Cresci-2017 and 0.95 on TwiBot-22, and a higher precision-recall value and better F1-scores. In addition, computational efficiency can be verified by the runtime analysis; case studies prove the framework’s strength when handling various situations of fraudulent behaviors. The given solution offers a scalable, reliable, and accurate methodology of detecting social media fraud based on the combination of sophisticated sequence modeling, realistic data augmentation, and hyperparameter optimization.