The rapid expansion of the Internet of Things (IoT) and the advancement of 5G technology require strong cybersecurity measures within IoT frameworks. Traditional security methods are insufficient due to the wide variety and large number of IoT devices and their limited computational capabilities. With 5G enabling faster data transmission, security risks have increased, making effective protective measures essential. Cross-Site Scripting (XSS) attacks present a significant threat to IoT security. In response, we have developed a new approach using Artificial Neural Networks (ANNs) to identify and prevent XSS breaches in IoT systems over 5G networks. We significantly improved our model’s predictive performance by using filter and wrapper feature selection methods. We validated our approach using two datasets, NF-ToN-IoT-v2 and Edge-IIoTset, ensuring its strength and adaptability across different IoT environments. For the NF-ToN-IoT-v2 dataset with filter feature selection, our Bilayered Neural Network (2 × 10) achieved the highest accuracy of 99.84%. For the Edge-IIoTset dataset with filtered feature selection, the Trilayered Neural Network (3 × 10) achieved the best accuracy of 99.79%. We used ANOVA tests to address the sensitivity of neural network performance to initial conditions, confirming statistically significant improvements in detection accuracy. The ANOVA results validated the enhancements across different feature selection methods, demonstrating the consistency and reliability of our approach. Our method demonstrates outstanding accuracy and robustness, highlighting its potential as a reliable solution for enhancing IoT security in the era of 5G networks.

Leaf disease detection plays a crucial role in modern agricultural management, enabling early intervention to minimize crop losses. This paper explores the application of the YOLOv10 model for detecting and classifying banana leaf conditions with high accuracy. A publicly available dataset of 938 images was used, categorized into five classes, namely Black-Sigatoka, Healthy-Leaf, Panama-Disease, Potassium-Deficiency, and Yellow-Sigatoka. The model achieved a mean Average Precision (mAP@0.5) of 88.85%, a precision of 91.22%, and a recall of 85.06%, demonstrating strong detection capabilities. These findings highlight the effectiveness of YOLOv10 in advancing automated disease detection, providing a reliable tool for precision agriculture. The model’s ability to accurately classify multiple leaf conditions can aid farmers in proactive disease management, ultimately enhancing crop health and sustainability.

The Internet of Things (IoT) now appears in each domain, from smart cities to home applications. The widespread use of IoT is making its security a real concern. The past few years have revealed an extraordinary increase in computer-intensive applications. Such applications always make huge volumes of data that demand severe latency-aware computational processing abilities. While edge computing is one of the attractive technologies for balancing severe latency-related problems, its deployment produces novel tasks. Edge computing is an innovative model distinguished mainly by its mobility support, geo-distributed process, low latency, and context awareness. However, recent edge computing developments have begun to explore novel IoT potentials that are leveraged from a security perspective. Methods depend upon artificial intelligence (AI) and its subgroups, machine learning (ML) and deep learning (DL), are generally employed to develop a safe Intrusion Detection System (IDS) for IoT. This study proposes a Hybrid Deep Learning-Based Intrusion Detection for Edge Computing Using Starfish Optimization Algorithm (HDLID-ECSOA) technique. The main goal of the HDLID-ECSOA technique is to provide intelligent edge computing in smart cities using advanced optimization models. Initially, the data pre-processing employs the min-max normalization to convert and standardize raw data to improve the efficiency of models. Furthermore, the dingo optimizer algorithm (DOA) technique detects and chooses the most relevant features from input data. Moreover, integrating a convolutional neural network and bidirectional gated recurrent unit with a cross-attention mechanism (CNN-BiGRU-CrAM) technique is implemented for the classification process. To enhance model performance, the starfish optimization algorithm (SFOA) is used for hyperparameter tuning to select the optimal parameters for improved accuracy. A comprehensive experimentation analysis of the HDLID-ECSOA model is performed under the Edge-IIoT and ToN-IoT datasets. The experimental validation of the HDLID-ECSOA model portrayed superior accuracy values of 99.35% and 99.33% over existing techniques under the dual dataset.