جامعة المعرفة
المصدر المعتمد للمعلومات البحثية والمخرجات هو قاعدة بيانات سكوبس (Elsevier Scopus) مع تدقيق بيانات الباحثين من عمادة البحث العلمي.

Research outputs

المخرجات البحثية

مقالات ومراجعات وأوراق مؤتمرات وفصول كتب، مفهرسة برقم DOI ومربوطة بباحثي الجامعة.

6 مُخرَج نتائج البحث عن «computational efficiency»

مقالوصول مفتوح202529استشهاد· Semantic Scholar

Optimizing IoT intrusion detection with cosine similarity based dataset balancing and hybrid deep learning

Scientific Reports · مجلد 15 (1) · 30939-30939

With IoT networks expected to exceed 29 billion connected devices by 2030, the risk of cyberattacks has never been higher. As more devices come online, the attack surface for hackers continues to expand, making cybersecurity a pressing concern. Intrusion Detection Systems (IDS) are essential for identifying and mitigating these threats in real-time. However, a significant challenge IDS faces is dealing with imbalanced datasets, where attack instances are significantly underrepresented compared to normal traffic. Training models on such skewed data leads to a bias toward majority-class patterns, reducing their ability to detect intrusions effectively. To address this issue, this work introduces CSMCR (Cosine Similarity-based Majority Class Reduction), a novel technique that selectively removes redundant majority-class samples while preserving dataset integrity. Unlike traditional approaches like SMOTE (oversampling) or random undersampling, CSMCR ensures that the retained majority instances remain diverse by analyzing feature-wise similarity. This prevents unnecessary data duplication and minimizes information loss. Additionally, we developed a hybrid deep learning model integrating RegNet and FBNet architectures to enhance feature extraction and classification performance. Experimental results on multiple IDS datasets confirm that balancing the dataset to a 1:1 ratio optimally prevents overfitting and improves model interpretability. The proposed model achieved an F1-score of 0.9758 on RT-IoT2022 and 0.9275 on UNSW Bot-IoT, outperforming SMOTE-based methods in accuracy and computational efficiency. Notably, CSMCR reduced training time by 53% compared to conventional oversampling techniques. Incremental training evaluations reveal that bias formation reduces performance beyond a 1:2 majority-to-minority ratio. These findings establish CSMCR as a robust, scalable, and computationally efficient IDS balancing strategy tailored for IoT network security.

IoT intrusion detectionclass imbalancecosine similaritydata balancing
مقالوصول مغلق202512استشهاد· Semantic Scholar

Consumer Technology in Task Offloading and Edge Resource Allocation: AIoT and Edge Computing for Next-Generation Communication

IEEE Transactions on Consumer Electronics · مجلد 71 (2) · 5356-5365

Artificial Intelligence of Things (AIoT) edge computing has emerged as a critical enabler for consumer technology, easing the load on distribution grid networks through enhanced data transmission and processing capabilities. The Internet of Things (IoT), a modern network paradigm, connects devices using sensors and communication mediums to enable seamless information exchange. However, the limited computational capacity of edge nodes poses challenges in optimizing AIoT edge resources compared to cloud computing. To address this, a cloud-edge three-layer framework for task offloading and edge resource allocation is proposed in the context of consumer technology and distribution grids. This model incorporates random tasks, limited resources, unequal processing power, and high latency requirements. It features two key phases: a resource auction using a multi-round iterative process and compute offloading managed by a Deep Reinforcement Learning (DRL) approach. Enhanced algorithms such as Double DQN and Dueling DQN are integrated into the job offloading process to improve performance. Simulations validate the convergence of the task offloading algorithm and demonstrate that the proposed methods significantly enhance computational efficiency and resource utilization for edge nodes. These advancements offer promising solutions for the dynamic needs of consumer technology and future communication networks.

task offloadingedge resource allocationAIoTedge computing
مقالوصول مفتوح20254استشهاد· Semantic Scholar

Unravelling handwriting images: deep neural models for dyslexia, dysgraphia, and other learning disabilities detections and classifications: a literature review

PeerJ Computer Science · مجلد 11 · e3296-e3296

Background As a crucial cognitive and motor function, handwriting can reveal dyslexia, dysgraphia, and other neurodevelopmental conditions. With the emergence of AI, deep learning (DL) models can analyze handwriting patterns for early identification and categorization of these conditions with high level of accuracy. Objective This study presents a literature review by examining DL models in classifying learning disabilities using handwriting data. It identifies models’ shortcomings and suggests future research directions, improving model accuracy, fairness, and usability. Methodology By employing the standard literature guidelines, the authors identified 24 studies from a pool of 502 studies. They extracted dataset details, methodologies, and key findings from the studies. Findings The findings indicate the dominance of convolutional neural networks in improving learning disabilities classification through the extraction of fine-grained spatial handwritten patterns. Vision transformers (ViTs) show potential for long-range contextual feature extraction to enhance the model’s generalization capabilities. Despite these advances, dataset size, model generalizability, computational efficiency, interpretability, and real-world deployment continue to be significant challenges. Conclusions The review findings underscore the transformative potential of DL-based handwriting analysis in educational and clinical settings, providing early detection and intervention in learning disabilities.

handwriting analysisdyslexiadysgraphialearning disabilities
مقالوصول مغلق20252استشهاد· Semantic Scholar

Cross-Platform Acceleration of Holographic Rendering in IoT Devices Using Open CL-Based Heterogeneous Computing

IEEE Transactions on Consumer Electronics · مجلد 71 (4) · 11522-11531

The integration of holographic technology into IoT-enabled consumer electronics offers transformative opportunities for immersive visualization and real-time interaction. However, the high computational demand of digital holographic reconstruction algorithms poses significant challenges for real-time deployment and cross-platform scalability, limiting their broader application in IoT ecosystems. To address these challenges, this paper presents an OpenCL-based heterogeneous acceleration framework that enhances the performance and portability of digital holographic reconstruction. By leveraging the parallel computing capabilities of CPU+GPU collaboration and data-parallel programming under the OpenCL architecture, the proposed method achieves significant improvements in computational efficiency. Experimental results across various hologram resolutions and GPU platforms demonstrate up to$54.2 \times $overall speedup and a parallel acceleration ratio of$94.7 \times $compared to CPU-only execution. Furthermore, the approach exhibits excellent scalability and cross-platform compatibility, making it well-suited for the real-time and embedded requirements of next-generation holographic IoT consumer devices.

digital holographic reconstructionholographic renderingOpenCLheterogeneous computing
مقالوصول مغلق2025

Accelerating Prostate Cancer Detection Through Histopathological Image Analysis Using Artificial Intelligence

Microscopy Research and Technique · مجلد 89 (5) · 687-704

Prostate cancer is a prevalent and serious health concern, ranking among the most frequently diagnosed cancers and a leading cause of cancer-related deaths in men worldwide. Early detection and accurate diagnosis are crucial for improving patient outcomes by limiting disease progression. Histopathological image analysis remains the gold standard for prostate cancer detection; however, manual interpretation is time-consuming and requires specialized expertise. To address these challenges, this study proposes a hybrid deep learning framework that combines an ensemble of transfer-learned CNNs (VGG-16, DenseNet-121, and AlexNet) with a fine-tuned Vision Transformer (ViT). The CNN ensemble extracts rich local features, while the ViT captures global contextual dependencies through a self-attention mechanism and a multilayer perceptron (MLP). Additionally, a cross-attention fusion (CAF) module integrates local and global features, and knowledge distillation (KD) enables a lightweight student network suitable for efficient clinical deployment. The study utilizes the publicly available PANDA dataset for training and testing. Preprocessing steps, including patch generation, gamma correction, and stain deconvolution, enhance image quality and feature representation. A comprehensive evaluation was conducted using standard performance metrics such as accuracy, true positive rate (TPR), true negative rate (TNR), precision, F1-score, false negative rate (FNR), and false positive rate (FPR). An ablation study confirmed the contribution of each module, highlighting the critical role of ensemble CNNs, CAF, and ViT in improving performance. Experimental results demonstrate that the proposed model outperforms conventional transfer learning models and existing state-of-the-art techniques, achieving 97.91% accuracy, along with significant improvements in TPR, TNR, and reduced FNR/FPR. The computational complexity, evaluated in terms of parameters, FLOPs, GPU memory, and inference time, indicates that the proposed model is more demanding than traditional CNNs. Nevertheless, the architecture strikes a practical balance between predictive accuracy and efficiency, making it suitable for real-world clinical applications. These findings underscore the potential of AI-powered hybrid models in expediting prostate cancer diagnosis and enabling timely intervention for improved patient outcomes.

prostate cancer detectionhistopathology image analysishybrid CNN-ViT modelcross-attention fusion
مقالوصول مفتوح20241استشهاد· Semantic Scholar

ERCO-Net: Enhancing Image Dehazing for Optimized Detail Retention

International Journal of Advanced Computer Science and Applications · مجلد 15 (10)

Image dehazing is a crucial preprocessing step in computer vision for enhancing image quality and enabling many downstream applications. However, existing methods often do not accurately restore hazy images while maintaining computational efficiency. To overcome this challenge, we propose ERCO-Net a new fusion framework that combines edge restriction and contextual optimization methods. By using boundary constraints, ERCO-Net extend the boundaries that help in protecting the edges and structures of an image. Contextual optimization impacts the final quality of the dehazed image by enhancing smoothness and coherence. We compare ERCO-Net with conventional approaches such as dark channel prior (DCP), All-in-one dehazing network (AoD), and Feature fusion attention network (FFA-Net). The comparative evaluation highlights the effectiveness of the proposed fusion method, providing significant improvement in image clarity, contrast, and colors. The combination of edge restriction and contextual optimization not only enhances the quality of dehazing but also decreases computational complexity, presenting a promising avenue for advancing image restoration techniques. The source code is available at https://github.com/FatimaAyub12/Image-Dehazing-.

image dehazingedge preservationboundary constraintsimage fusion