检索结果-内蒙古大学图书馆

Robust speech perception and classification-driven deep convolutional neural network with natural language processing

引用

Alexandria engineering Journal 2025年 123卷 358-368页

作者： Alshammari, Alya Alzaidi, Muhammad Swaileh A. Alrusaini, Othman Alshuhail, Asma Allafi, Randa Yafoz, Ayman Alsini, Raed Khadidos, Alaa O. Department of Applied Linguistics College of Languages Princess Nourah bint Abdulrahman University P.O. Box 84428 Riyadh11671 Saudi Arabia Department of English Language College of Language Sciences King Saud University P. O. Box 145111 Riyadh Saudi Arabia Department of Engineering and Applied Sciences Applied College Umm Al-Qura University Makkah Saudi Arabia Department of Information Systems College of Computer Sciences & Information Technology King Faisal University Saudi Arabia Department of Computer Science College of Sciences Northern Border University Arar Saudi Arabia Department of Information Systems Faculty of Computing and Information Technology King Abdulaziz University Jeddah Saudi Arabia

Speech recognition systems use computer algorithms to interpret and process spoken words and translate them into text. Speech-to-text, or Speech detection, is the proficiency of a program or machine to find spoken words aloud and translate them into plain text. Speech detection systems are the most common complex systems that use wide-ranging research in computer engineering, linguistics, and computer science. In today's significant data era, several machine learning (ML) approaches for processing raw, unlabelled speech data still need to be improved. Deep learning (DL) models, on the other hand, with their powerful modelling ability for extensive data, are capable of directly processing unlabelled data and have become a crucial area of research in the domain of automated speech recognition. The DL model has recently gained popularity for voice disorders and automated speech detection. DL is an ML that exploits artificial neural networks (ANN) to gain meaningful insights from data. The DL approach demonstrated great success and achieved significant attention in different fields comprising speech processing, computer vision (CV), and natural language processing (NLP). Therefore, this article develops a Deep Learning with Natural Language Processing based Robust Speech Detection (DLNLP-RSD) technique. The DLNLP-RSD technique derives features from the mel-frequency cepstral coefficient (MFCC) technique. Followed by the DLNLP-RSD technique employs a one-dimensional deep convolutional neural network (CNN) approach for the detection and classification of speech. Moreover, the parameter tuning model uses the grasshopper fractals optimization algorithm (GOA) approach. The performance evaluation of the proposed technique highlights its significant performance over other models concerning distinct measures. An extensive simulation analysis is executed to examine the improved results of the DLNLP-RSD model. The experimental validation of the DLNLP-RSD technique portrayed a sup

关键词： Convolutional neural networks

来源：评论

学校读者我要写书评

暂无评论

FPGA-Optimized Neuromorphic Modeling of Cardiac Purkinje Fibers for Next-Generation Bionic Implants

IEEE Transactions on Medical Robotics and Bionics

引用

IEEE Transactions on Medical Robotics and Bionics 2025年

作者： Ghanbarpour, Gilda Chaudhary, Muhammad Akmal Assaad, Maher Ghanbarpour, Milad Razi University Department of Computer Engineering Kermanshah Iran Ajman University Department of Electrical and Computer Engineering College of Engineering and Information Technology United Arab Emirates Kermanshah University of Technology Electrical Faculty Department of Electrical Engineering Kermanshah Iran

The optimized hardware implementation of neurons and biological cells in the neuromorphic domain is of significant importance. In this paper, a novel method is presented that reduces any number of nonlinear terms in the differential equations describing the behavior of neurons or biological cells with a common variable to a single nonlinear term with high precision. This approach significantly improves implementation efficiency by reducing hardware resource consumption while maintaining high frequency and accuracy. The proposed method was applied to Cardiac Purkinje Fiber Cells, and its validity was demonstrated through time-domain analysis, noise condition analysis, Lyapunov stability analysis, and bifurcation analysis to validate the model under various conditions. These validations ensure the accuracy and stability of the proposed approach across different operating conditions. To assess large-scale applicability, the model was tested in a 300-cell Purkinje fiber network, demonstrating accurate synchronization, equilibrium states, and cross-spectral consistency while maintaining computational efficiency. The digital hardware implementation on a Virtex-7 FPGA board demonstrated a frequency improvement of 3.49 times compared to the original model and 1.79 times compared to the best implementation of this model to date. We also simulated a network of 4500 cells to analyze correlation and implemented it on hardware to demonstrate that the proposed model, based on the method presented in this paper, can efficiently and accurately scale to large-scale applications. This efficient and scalable approach paves the way for applications in medical research, bioengineering, and neuromorphic hardware development, including the creation of hardware-accelerated tools for simulating biological systems, designing bio-inspired devices, and enabling large-scale real-time simulations for understanding and treating cardiac or neurological conditions. © 2018 IEEE.

关键词： Neurons

来源：评论

学校读者我要写书评

暂无评论

Application of an innovative FPGA-optimized model on pancreatic β-cells: Validation via Lyapunov analysis and equilibrium stability

引用

AEU - International Journal of Electronics and Communications 2025年 197卷

作者： Ghanbarpour, Gilda Chaudhary, Muhammad Akmal Assaad, Maher Ghanbarpour, Milad Department of Computer Engineering Razi University Kermanshah Iran Department of Electrical and Computer Engineering College of Engineering and Information Technology Ajman University United Arab Emirates Department of Electrical Engineering Energy Faculty Kermanshah University of Technology Kermanshah Iran

Presenting an ideal hardware model capable of mimicking behavior of biological cells and neurons holds great significance in neuromorphic engineering. This research introduces a novel and unique approach to digitally implementing a biological cell model with the goal of reducing hardware requirements and enhancing processing speed while maintaining signal accuracy. The method outlined focuses on minimizing the number of nonlinear terms in the original model, making it well-suited for digital implementation. Specifically, this approach is applied to the pancreatic β cell model in this study. Both software simulation and hardware implementation results validate the effectiveness of the proposed method in reproducing the signal and behavior of the original β cell model by decreasing the number of nonlinear terms without discarding any equations from the initial model. By consolidating them into a single form, this method reduces the number of non-linear terms to one through a common variable, making it highly suitable for digital implementation. To validate the mathematical simulation findings, the proposed model was synthesized and implemented on the Zynq XC7Z010 (3CLG400) reconfigurable board (FPGA). The results demonstrate the ability to more efficiently reproduce biological behavior at significantly lower execution costs. Implementing this technique on the Zynq board can enhance the speed of the suggested model by up to 2.187 times compared to the original model, while also achieving a 34.91% decrease in energy (power) usage. © 2025 Elsevier GmbH

关键词： Cell culture

来源：评论

学校读者我要写书评

暂无评论

High-Voltage SF6 Circuit Breaker Maintenance 33rd

High-Voltage SF6 Circuit Breaker Maintenance

引用

33rd International Conference on Organization and technology of Maintenance, OTO 2024

作者： Soldo, Luka Knežević, Goran Zovko, Mario Croatian Transmission System Operator Transmission Area Osijek Vukovarska Cesta 217 Osijek31000 Croatia Faculty of Electrical Engineering Computer Science and Information Technology Osijek Kneza Trpmira 2b Osijek31000 Croatia

ISBN: (纸本)9783031805967

Nowadays, the need for uninterrupted power supply is greater than ever. One of the most effective ways to ensure an uninterrupted supply of electricity is the correct selection of the method and interval of maintenance of elements in a high-voltage substation. In this way, not only is a high-quality delivery of electricity ensured, but also the possibility of equipment failure that can be dangerous for people in the substation is reduced. A high-voltage circuit breaker is a device that performs several different functions. It is used for network configuration changes, and also provides protection for equipment and personnel. High voltages and thermal (current) stresses cause aging and wear of switch parts, and through various forms of maintenance and testing, damage parts are determined and potential hazards are safely removed. This paper explains the maintenance procedures of the high-voltage SF6 circuit breaker. Using the example of a high-voltage line-bay circuit breaker, the entire 6Y circuit breaker revision procedure (revision that is carried out in six-year cycles) is presented. In addition to visual inspection, checking, cleaning and replacement of worn parts, the switch on and off times were recorded. The quality of SF6 gas was also tested by measuring the concentration of SF6 gas decomposition products and the concentration of mineral oil aerosols in SF6 gas, by measuring the dew point of the gas and controlling the volume percentage of SF6 gas. The voltage drops at the primary connections of the switches were measured, and the operation of the pressure relays was also checked. © The Author(s), under exclusive license to Springer Nature Switzerland AG 2025.

关键词： Mineral oils

来源：评论

学校读者我要写书评

暂无评论

Multilingual Hate Speech Detection: Innovations in Optimized Deep Learning for English and Arabic Hate Speech Detection

引用

SN computer Science 2025年第3期6卷 1-16页

作者： AL-Sukhani, Hassan Bsoul, Qusay Elhawary, Abdelrahman H. Nasr, Ziad M. Mansour, Ahmed E. Batyha, Radwan M. Alqadi, Basma S. Alqurni, Jehad Saad Alfagham, Hayat Madbouly, Magda M. Cybersecurity Department Faculty of Science and Information Technology Jadara University Irbid Jordan Cybersecurity Department College of Computer Sciences and Informatics Amman Arab University Amman Jordan Faculty of Engineering Ain Shams University Cairo Egypt Faculty of Computer Science Misr International University Cairo Egypt Faculty of Information Technology Applied Science Private University Amman 11931 Jordan Computer Science Department College of Computer and Information Sciences Al Imam Mohammad Ibn Saud Islamic University Riyadh Saudi Arabia Department of Educational Technologies College of Education Imam Abdulrahman Bin Faisal University P.O. Box 1982 Dammam Saudi Arabia Department of MIS College of Applied Studies and Community Service Imam Abdulrahman Bin Faisal University P.O. Box 1982 Dammam Saudi Arabia Institute of Graduate Studies & Research Alexandria University Alexandria Egypt Faculty of Computer and Data Science Alexandria University Alexandria Egypt

This paper presents the development of a multilingual hate speech detection model that effectively processes and classifies content in both Arabic and English. The study leverages both traditional machine learning models, such as K-Nearest Neighbors (KNN), Naive Bayes, and Support Vector Machines (SVM), as well as advanced deep learning models, specifically Bi-directional Long Short-Term Memory (Bi-LSTM) networks. A key challenge addressed is the classification of mixed-language content, which is common on social media platforms in the MENA region. To enhance detection performance, preprocessing techniques were applied to the text data, and the Synthetic Minority Over-sampling Technique (SMOTE) was used to balance the dataset. The results show that the Bi-LSTM model outperformed traditional machine learning approaches, particularly in identifying hate speech across multiple languages. The proposed model demonstrates superior accuracy and robustness in handling mixed-language input, providing a more effective solution for real-world hate speech detection tasks. © The Author(s), under exclusive licence to Springer Nature Singapore Pte Ltd. 2025.

关键词： Class imbalance Deep learning Hate speech detection Machine learning Mixed Arabic and English training Multilingual NLP Natural language processing (NLP) Sentiment analysis

来源：评论

学校读者我要写书评

暂无评论

Adaptive Cooperative Quality Weight Spectrum Sensing for Mitigating Byzantine Attacks in Cognitive Radio

引用

IEEE Access 2025年 13卷 80930-80940页

作者： Srisomboon, Kanabadee Pipanmekaporn, Luepol Prayote, Akara Lee, Wilaiporn Faculty of Engineering King Mongkut’s University of Technology Communication and Computer Network Research Group (C2NRG) Department of Electrical and Computer Engineering North Bangkok Thailand Faculty of Applied Science King Mongkut’s University of Technology Department of Computer and Information Science North Bangkok Thailand

Spectrum sensing data falsification (SSDF) attack, i.e., Byzantine attack, is one of the critical threats of the cooperative spectrum sensing where the Byzantine attackers (BAs) forward incorrect local sensing results to mislead the fusion center on channel availability decisions. By using traditional voting rule, the cooperative spectrum sensing performance deteriorates significantly due to incorrect local sensing results. Then, reliability weight strategy becomes the popular solution to avoid incorrect sensing results from BAs and unreliable cognitive radio users (CRUs). However, it is very difficult to detect the attackers since they also occasionally provide correct sensing results to the fusion center for concealing the attack objective. Based on existing techniques, the BAs and CRUs may be assigned with low reliability weights or distinguished from the data fusion account. However, it is very difficult to detect the attackers since they also occasionally provide correct sensing results to the fusion center for concealing the attack objective. Then, existing techniques still suffer from BAs and negative impact of unreliable CRUs. In this paper, we propose the adaptive cooperative quality weight algorithm for mitigating the Byzantine attack issue by distinguishing the BAs and CRUs from the data fusion account while selecting only useful CRUs since the number of members in the account is also the important factor for cooperative spectrum sensing. In our proposed algorithm, we adopt a stable preference ordering towards ideal solution (SPOTIS) for determining the reliability of SUs which shows low computational complexity as compared to other reliability weight-based techniques. To achieve high sensing performance, our global decision threshold is adapted according to the reliability of reliable users. From the simulation results, our proposed algorithm significantly improves global detection probability and total error probability compared to the traditional votin

关键词： Adaptive Thresholding Byzantine Attack Cooperative Spectrum Sensing CR SPOTIS

来源：评论

学校读者我要写书评

暂无评论

Multimodal Representation Learning Based on Personalized Graph-Based Fusion for Mortality Prediction Using Electronic Medical Records

引用

Big Data Mining and Analytics 2025年第4期8卷 933-950页

作者： Abdulrahman Al-Dailami Hulin Kuang Jianxin Wang Hunan Provincial Key Lab on Bioinformatics School of Computer Science and Engineering Central South University Changsha China Faculty of Computer and Information Technology Sana'a University Sana'a Yemen

Predicting mortality risk in the Intensive Care Unit (ICU) using Electronic Medical Records (EMR) is crucial for identifying patients in need of immediate attention. However, the incompleteness and the variability of EMR features for each patient make mortality prediction challenging. This study proposes a multimodal representation learning framework based on a novel personalized graph-based fusion approach to address these challenges. The proposed approach involves constructing patient-specific modality aggregation graphs to provide information about the features associated with each patient from incomplete multimodal data, enabling the effective and explainable fusion of the incomplete features. Modality-specific encoders are employed to encode each modality feature separately. To tackle the variability and incompleteness of input features among patients, a novel personalized graph-based fusion method is proposed to fuse patient-specific multimodal feature representations based on the constructed modality aggregation graphs. Furthermore, a MultiModal Gated Contrastive Representation Learning (MMGCRL) method is proposed to facilitate capturing adequate complementary information from multimodal representations and improve model performance. We evaluate the proposed framework using the large-scale ICU dataset, MIMIC-III. Experimental results demonstrate its effectiveness in mortality prediction, outperforming several state-of-the-art methods.

关键词： Representation learning Accuracy MIMICs Mortality Logic gates Big Data Graph neural networks Data mining Electronic medical records

来源：评论

学校读者我要写书评

暂无评论

Guest Editorial: Special Issue of Selected Extended Papers From the IEEE International Conference on Flexible, Printable Sensors and Systems (FLEPS) 2024

IEEE Journal on Flexible Electronics

引用

IEEE Journal on Flexible Electronics 2025年第1期4卷 2-3页

作者： Mantysalo, Matti Marrocco, Gaetano Agarwale, Shweta Grau, Gerd Gao, Wei Tampere University Faculty of Information Technology and Communication Sciences Tampere33720 Finland University of Roma Tor Vergata Dipartimento di Ingegneria Civile ed ed Ingegneria Informatica Rome00133 Italy Aarhus University Department of Electrical and Computer Engineering Aarhus8200 Denmark York University Department of Electrical Engineering and Computer Science TorontoONM3J 1P3 Canada Caltech Department of Engineering and Applied Science PasadenaCA91125 United States

来源：评论

学校读者我要写书评

暂无评论

ResTransUNet: A hybrid CNN-transformer approach for liver and tumor segmentation in CT images

引用

computers in Biology and Medicine 2025年 190卷 110048-110048页

作者： Anwar, Asmaa Sabet Amin, Khaled Hadhoud, Mohiy M. Ibrahim, Mina Department of Computer Engineering Faculty of Engineering May University Cairo Egypt Department of Information Technology Faculty of Computers and Information Menoufia University Shebin Elkom Egypt Department of Machine Intelligence Faculty of Artificial Intelligence Menoufia University Shebin Elkom Egypt

Background and Objective: Accurate medical tumor segmentation is critical for early diagnosis and treatment planning, significantly improving patient outcomes. This study aims to enhance liver and tumor segmentation from CT and liver images by developing a novel model, ResTransUNet, which combines convolutional and transformer blocks to improve segmentation accuracy. Methods: The proposed ResTransUNet model is a custom implementation inspired by the TransUNet architecture, featuring a Standalone Transformer Block and ResNet50 as the backbone for the encoder. The hybrid architecture leverages the strengths of Convolutional Neural Networks (CNNs) and Transformer blocks to capture both local features and global context effectively. The encoder utilizes a pre-trained ResNet50 to extract rich hierarchical features, with key feature maps to preserved it as skip connections. The Standalone Transformer Block, integrated into the model, employs multi-head attention mechanisms to capture long-range dependencies across the image, enhancing segmentation performance in complex cases. The decoder reconstructs the segmentation mask by progressively upsampling encoded features while integrating skip connections, ensuring both semantic information and spatial details are retained. This process culminates in a precise binary segmentation mask that effectively distinguishes liver and tumor regions. Results: The ResTransUNet model achieved superior Dice Similarity Coefficient (DSC) for liver segmentation (98.3% on LiTS and 98.4% on 3D-IRCADb-01) and for tumor segmentation from CT images (94.7% on LiTS and 89.8% on 3D-IRCADb-01) as well as from liver images (94.6% on LiTS and 91.1% on 3D-IRCADb-01). The model also demonstrated high precision, sensitivity, and specificity, outperforming current state-of-the-art methods in these tasks. Conclusions: The ResTransUNet model demonstrates robust and accurate performance in complex medical image segmentation tasks, particularly in liver and tum

关键词： computerized tomography

来源：评论

学校读者我要写书评

暂无评论

DSTNet: Distinguishing Source and Target Areas for Image Copy-Move Forgery Detection 27th

DSTNet: Distinguishing Source and Target Areas for Image Cop...

引用

27th International Conference on Pattern Recognition, ICPR 2024

作者： Zhao, Kaiqi Yuan, Xiaochen Huang, Guoheng Liu, Kun School of Cyberspace Security Shandong University of Political Science and Law Jinan China Faculty of Applied Sciences Macao Polytechnic University China School of Computer Science and Technology Guangdong University of Technology Guangzhou China School of Information Science and Engineering University of Jinan Jinan China

ISBN: (纸本)9783031783111

In copy-move forgery detection, most relevant studies concern locating the copy-move areas without the distinction of source and target regions. This paper proposes an end-to-end network, DSTNet, to identify the source and target based on consistency detection between the copy-move region and the non-copy-move region. The DSTNet is composed of two stages, the Pre-processing stage and the Discrimination stage. Pre-processing Stage extracts internal information of copy-move and non-copy-move areas and conducts a series of operations to meet the requirements of network input. Discrimination stage allows multiple patches for input and classifies the input patches. Specifically, the Pre-processing stage, contains the Copy-move Patches Selection (CM Patches Selection) and Genuine Patches Selection, can select pairs of copy-move and none copy-move patches. We train the proposed DSTNet on two large synthetic datasets and use the public datasets CASIA and Comofod for evaluation. The experiment shows that our method achieves excellent results. Particularly, we achieve a 5.4% higher F1 based on ground-truth of copy-move mask (GT-CM) on CASIA dataset. © The Author(s), under exclusive license to Springer Nature Switzerland AG 2025.

关键词： Large datasets

来源：评论

学校读者我要写书评

暂无评论

建议与咨询留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

时间限定

文献类型

馆藏选择

核心期刊

语言

文献类型

帮助

文字说明：

检索规则说明：

检索范例：

分类表

所选分类

限定检索结果

文献类型

馆藏范围

日期分布

学科分类号

主题

机构

作者

语言

请选择保存的检索档案：

请选择收藏分类：

通借通还

建议与咨询 留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

时间限定

文献类型

馆藏选择

核心期刊

语言

文献类型

帮助

文字说明：

检索规则说明：

检索范例：

分类表

所选分类

限定检索结果

文献类型

馆藏范围

日期分布

学科分类号

主题

机构

作者

语言

请选择保存的检索档案： 新增检索档案 确定 取消

请选择收藏分类： 新增自定义分类 确定 取消

通借通还

建议与咨询留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

请选择保存的检索档案：

请选择收藏分类：