Article
Computer Science, Information Systems
Yingjie Tian, Weizhi Gao, Qin Zhang, Pu Sun, Dongkuan Xu
Summary: Image classification is crucial in the IoT system, and long-tailed distribution data is prevalent in our daily lives. The imbalanced classes in long-tailed classification lead to a significant performance gap between training and testing. To address this issue, a novel class-based covariance transfer method is proposed to properly transfer the covariance information in long-tailed classification tasks.
INTERNET OF THINGS
(2023)
Article
Mathematics
Abdelghani Dahou, Samia Allaoua Chelloug, Mai Alduailij, Mohamed Abd Elaziz
Summary: This work proposes a framework to address data analysis and performance improvement in the Social Internet of Things (SIoT) system. The framework consists of a deep learning model for feature extraction and an optimization algorithm for feature selection. Experimental results demonstrate that the developed method outperforms other models in the SIoT environment.
Article
Computer Science, Information Systems
Zhihan Lv, Ranran Lou, Amit Kumar Singh, Qingjun Wang
Summary: This study investigates the data classification and resource optimization for the Internet of Things in the context of 5G communication network. A virtualization architecture and an improved classification algorithm are designed, leading to significant improvements in resource utilization efficiency.
ACM TRANSACTIONS ON INTERNET TECHNOLOGY
(2022)
Article
Computer Science, Artificial Intelligence
Taher M. Ghazal, Nasser Taleb
Summary: Ovarian cancer is a difficult-to-diagnose tumor in its early stages, but this research proposes a new method for distinguishing it from other cancers. By optimizing neural network configuration, the detection of ovarian cancer is improved with high precision and low root mean square error. SOM and NN techniques show excellent performance in the identification and accuracy of ovarian cancer.
Article
Computer Science, Information Systems
Zhiwei Guo, Yu Shen, Shaohua Wan, Wen-Long Shang, Keping Yu
Summary: In this paper, a hybrid intelligence-driven medical image recognition framework combining deep learning with conventional machine learning is proposed to solve the issue of remote patient diagnosis in smart cities. Experimental results reveal that the framework improves recognition accuracy by approximately two to three percent compared to traditional methods.
IEEE JOURNAL OF BIOMEDICAL AND HEALTH INFORMATICS
(2022)
Article
Computer Science, Information Systems
Wenhan Liu, Qianxi Guo, Xinwei Gao, Sheng Chang, Hao Wang, Jin He, Qijun Huang
Summary: This article proposes a new unsupervised feature learning method for processing unlabeled 12-lead electrocardiogram signals. The method takes into account the characteristics of 12-lead ECGs and utilizes lead separation and combination to learn feature representations. Experimental results demonstrate that the method achieves good accuracy in myocardial infarction and atrial fibrillation detection.
IEEE INTERNET OF THINGS JOURNAL
(2022)
Article
Engineering, Electrical & Electronic
Yongzhao Xu, Gabriel Holanda, Luis Fabricio de F. Souza, Hercules Silva, Adriell Gomes, Iagson Silva, Marcos Ferreira Jr, Chuanyu Jia, Tao Han, Victor Hugo C. de Albuquerque, Pedro P. Reboucas Filho
Summary: This study proposes a fully automatic system based on Health of Things for classifying CT images of the skull through deep learning networks, and segmenting strokes through a learning transfer process combined with machine learning methods. The innovative method achieved excellent results in both classification and segmentation, surpassing literature methods based on automatic models.
IEEE SENSORS JOURNAL
(2021)
Article
Engineering, Multidisciplinary
Mohammed Al-Jabbar, Ebtesam Al-Mansor, S. Abdel-Khalek, Salem Alkhalaf
Summary: This study proposes a technique for effective scene classification and intrusion detection of remote sensing images in the IoT environment using deep learning. The technique involves a two-stage process, with the first stage using a modified DarkNet-53 feature extractor, EOA-based hyperparameter tuning, and graph convolution network (GCN) based classification for scene classification, and the second stage employing variational autoencoder (VAE) based intrusion detection. Experimental results demonstrate the improved performance of the technique in scene classification and intrusion detection.
ALEXANDRIA ENGINEERING JOURNAL
(2023)
Article
Computer Science, Artificial Intelligence
Zhenbing Liu, Fengfeng Wu, Yumeng Wang, Mengyu Yang, Xipeng Pan
Summary: In this study, a federated contrastive learning (FedCL) approach is proposed for training a shared deep learning model with privacy protection in distributed medical institutions. By integrating the idea of contrastive learning into the federated learning framework, FedCL combines the local model and the global model for contrastive learning, improving the generalization ability of the model. Experimental results on two public datasets demonstrate that our method outperforms other federated learning algorithms in medical image classification.
PATTERN RECOGNITION
(2023)
Article
Computer Science, Artificial Intelligence
Rishav Singh, Vandana Bharti, Vishal Purohit, Abhinav Kumar, Amit Kumar Singh, Sanjay Kumar Singh
Summary: The study addresses the challenges of long-tailed distributions and lack of high-quality annotated images in medical datasets. By formulating a few-shot learning problem and proposing a meta-learning-based MetaMed approach, the model achieved promising results with an accuracy of over 70% on three medical datasets, showcasing improved generalization capability.
PATTERN RECOGNITION
(2021)
Article
Computer Science, Artificial Intelligence
Benteng Ma, Yu Feng, Geng Chen, Changyang Li, Yong Xia
Summary: Medical data sharing is crucial but suffers from privacy issues. This paper proposes a novel federated learning algorithm, FedAR, which addresses data heterogeneity by employing a flexible re-weighting scheme and achieves superior performance.
PATTERN RECOGNITION
(2023)
Article
Computer Science, Artificial Intelligence
A. Ponmalar, V Dhanakoti
Summary: This paper presents a novel technique to enhance intrusion detection by addressing the complexities of heterogeneous security data in big data. The proposed methodology significantly improves accuracy and can identify different types of attacks. Comparisons with baseline models demonstrate the effectiveness of the approach.
APPLIED SOFT COMPUTING
(2022)
Article
Computer Science, Information Systems
Xinyu Tong, Ziao Yu, Xiaohua Tian, Houdong Ge, Xinbing Wang
Summary: This paper proposes a machine learning based method to improve the accuracy of automated optical inspection (AOI). By utilizing a customized deep learning model to preprocess and classify the images taken by AOI machine, the method reduces the rate of misjudgment significantly. Experimental results demonstrate the importance of this method for practical production of PCBs.
FRONTIERS OF COMPUTER SCIENCE
(2022)
Article
Chemistry, Analytical
Ali Walid Daher, Ali Rizik, Marco Muselli, Hussein Chible, Daniele D. Caviglia
Summary: Edge Computing allows for measurement and cognitive decisions to be made outside a central server by utilizing data storage, manipulation, and processing on IoT nodes. The Raspberry Pi is a low-cost computing platform profitably applied in the IoT field, while Rulex is identified as a suitable ML platform for implementation on the Raspberry Pi.
Article
Computer Science, Artificial Intelligence
Taslim Murad, Sarwan Ali, Imdadullah Khan, Murray Patterson
Summary: This paper proposes a deep learning-based method called Spike2CGR, which converts spike sequences of the coronavirus into image representations for lineage and host classification tasks. The authors also design modified versions of Spike2CGR based on biological knowledge, which outperform the state-of-the-art method in terms of predictive performance. Experimental results demonstrate that Spike2CGR achieves better predictive performance for variant and host classification on spike sequences.
Article
Computer Science, Interdisciplinary Applications
Mohamed Abd Elaziz, Mohammed A. A. Al-qaness, Abdelghani Dahou, Rehab Ali Ibrahim, Ahmed A. Abd El-Latif
Summary: This paper proposes an efficient intrusion detection system for IoT-cloud based environments, using swarm intelligence algorithms and deep neural networks. Deep neural networks are used to obtain optimal features from IoT IDS data, and a feature selection technique based on the Capuchin Search Algorithm (CapSA) is proposed. The developed model, CNN-CapSA, is tested with four IoT-Cloud datasets and compared with other optimization algorithms, showing competitive performance.
ADVANCES IN ENGINEERING SOFTWARE
(2023)
Article
Environmental Sciences
Mohammed A. A. Al-qaness, Ahmed A. Ewees, Hung Vo Thanh, Ayman Mutahar AlRassas, Abdelghani Dahou, Mohamed Abd Elaziz
Summary: Decreasing fossil fuel utilization and anthropogenic greenhouse gases is a global goal to combat climate change and air pollution. Underground carbon storage (UCS) is a promising solution, but there are barriers to its global application. In this study, a hybrid algorithm called AOSMA was developed using swarm intelligence to enhance the prediction capability of the LSTM model. Evaluation experiments showed that AOSMA outperformed other algorithms in predicting CO2 storage efficiencies.
ENVIRONMENTAL SCIENCE AND POLLUTION RESEARCH
(2023)
Article
Automation & Control Systems
Mohammed A. A. Al-qaness, Abdelghani Dahou, Mohamed Abd Elaziz, A. M. Helmi
Summary: This study developed a deep learning architecture based on a multi-level residual network for human activity recognition, which improved recognition accuracy through model integration and feature extraction, and was evaluated and compared on multiple datasets, showing significant performance.
IEEE TRANSACTIONS ON INDUSTRIAL INFORMATICS
(2023)
Article
Mathematics
Mohammed A. A. Al-qaness, Abdelghani Dahou, Ahmed A. A. Ewees, Laith Abualigah, Jianzhu Huai, Mohamed Abd Elaziz, Ahmed M. M. Helmi
Summary: Many Chinese cities suffer from severe air pollution due to rapid economic development, urbanization, and industrialization. Particulate matter (PM2.5) is a major component of air pollutants and is associated with cardiopulmonary and other systemic diseases due to its ability to penetrate the human respiratory system. Forecasting PM2.5 concentration is vital for governments and local authorities to plan and take necessary actions.
Article
Mathematics
Mohamed Abd Elaziz, Abdelghani Dahou, Dina Ahmed Orabi, Samah Alshathri, Eman M. Soliman, Ahmed A. Ewees
Summary: The rapid spread of fake information and news related to the COVID-19 pandemic on social media platforms has raised serious concerns for public health and safety. This paper proposes a disinformation detection framework using multi-task learning and meta-heuristic algorithms to analyze Arabic social media posts. The experimental results show that the proposed framework achieves an accuracy of 59% and outperforms other algorithms in all evaluation measures.
Review
Mathematics, Interdisciplinary Applications
Essraa Gamal Mohamed, Rebeca P. Diaz Redondo, Abdelrahim Koura, Mohamed Sherif EL-Mofty, Mohammed Kayed
Summary: The significance of age estimation lies in its applications in various fields such as forensics, criminal investigation, and illegal immigration. With the increasing importance of age estimation, more investigation and development in this area of study are required. Several methods, including the utilization of biometric traits such as the face, teeth, bones, and voice, have been employed for age estimation. This paper summarizes the common biometric traits for age estimation and examines the use of this information in previous research, with a specific focus on traditional machine learning methods and deep learning approaches used for dental age estimation. The advances in convolutional neural network (CNN) models for dental age estimation from radiological images, such as 3D cone-beam computed tomography (CBCT), X-ray, and orthopantomography (OPG), are also discussed. Furthermore, potential innovations that could enhance the performance of age estimation systems are highlighted.
Article
Chemistry, Analytical
Abdulaziz Fatani, Abdelghani Dahou, Mohamed Abd Elaziz, Mohammed A. A. Al-qaness, Songfeng Lu, Saad Ali Alfadhli, Shayem Saleh Alresheedi
Summary: Intrusion detection systems (IDS) are vital for network security and identifying malicious activity. Both metaheuristic optimization algorithms and deep learning techniques have been used to enhance the accuracy and efficiency of IDS. This paper proposes a new IDS model that combines deep learning and optimization methods. The model incorporates a CNN-based feature extraction method and a modified version of the Growth Optimizer (GO) called MGO for feature selection. The Whale Optimization Algorithm (WOA) is employed to improve the search process. Extensive evaluation on public datasets of cloud and IoT environments demonstrates promising results, with the MGO outperforming previous methods in all experimental comparisons.
Article
Medicine, General & Internal
Abdelghani Dahou, Ahmad O. Aseeri, Alhassan Mabrouk, Rehab Ali Ibrahim, Mohammed Azmi Al-Betar, Mohamed Abd Elaziz
Summary: In this paper, a robust skin cancer detection framework is proposed to improve the accuracy by extracting and learning relevant image representations using a MobileNetV3 architecture. The modified Hunger Games Search (HGS) based on Particle Swarm Optimization (PSO) and Dynamic-Opposite Learning (DOLHGS) is used as a novel feature selection to maximize the model's performance. Experimental results show that the proposed approach outperforms other well-known algorithms in terms of classification accuracy and optimized features.
Article
Medicine, General & Internal
Mohamed Abd Elaziz, Abdelghani Dahou, Alhassan Mabrouk, Rehab Ali Ibrahim, Ahmad O. Aseeri
Summary: This paper proposes a framework that integrates deep learning and optimization techniques to improve prediction accuracy and provide real-time medical diagnosis in the 6G-enabled IoMT. The framework preprocesses medical computed tomography images, extracts features using a neural network, and applies an optimized algorithm to enhance classification performance. Evaluation experiments demonstrate the remarkable performance of this framework on multiple datasets.
Article
Computer Science, Artificial Intelligence
Alhassan Mabrouk, Rebeca P. Diaz Redondo, Mohamed Abd Elaziz, Mohammed Kayed
Summary: Federated learning is a convenient approach for scenarios with privacy concerns and quick reaction demands. This paper proposes an ensemble federated learning approach for privacy protection. Different computation nodes use different datasets and ensemble models for local computation, and the models are aggregated to obtain a global model. Experimental results show that the proposed method outperforms centralized approaches in Chest X-ray images.
APPLIED SOFT COMPUTING
(2023)
Article
Green & Sustainable Science & Technology
Yunes Almansoub, Ming Zhong, Muhammad Safdar, Asif Raza, Abdelghani Dahou, Mohammed A. A. Al-qaness
Summary: This study analyzes the impact of transportation infrastructure-based accessibility on mixed land use (MLU) patterns using deep neural network models. The study finds that areas close to the city center in the Jiang'an District of Wuhan show a higher prevalence of MLU, especially in regions with good accessibility to non-motorized and public transit options. The study also shows that transportation supply significantly influences MLU patterns.
Article
Business
Abdelghani Dahou, Alhassan Mabrouk, Ahmed A. Ewees, Marwa A. Gaheen, Mohamed Abd Elaziz
Summary: Social media can be used to inform the public about crises and emergencies, but it is often filled with irrelevant information. Researchers have focused on developing event detection systems to extract relevant events and their types, using deep learning techniques. This paper proposes a new event detection model that combines the MobileBERT model and a novel feature selection method to improve performance. Experiments conducted using real-world datasets show that the modified feature selection method enhances the performance of the proposed framework for event detection tasks compared to existing methods.
TECHNOLOGICAL FORECASTING AND SOCIAL CHANGE
(2023)