检索结果-内蒙古大学图书馆

Investigation of attention mechanism for speech command recognition

Multimedia Tools and Applications 2024年 1-16页

作者： Xie, Jie Zhu, Mingying Hu, Kai Zhang, Jinglan Guo, Ya School of Computer and Electronic Information and School of Artificial Intelligence Nanjing Normal University Nanjing China School of Economics Nanjing University 22 Hankou RoadJiangsu Nanjing210093 China Johns Hopkins University - Hopkins-Nanjing Center for Chinese and American Studies Nanjing University 162 Shanghai RoadJiangsu Nanjing210093 China School of Internet of Things Engineering Jiangnan University Wuxi China School of Computer Science Faculty of Science Queensland University of Technology Brisbane Australia

As an application area of speech command recognition, the smart home has provided people with a convenient way to communicate with various digital devices. Deep learning has demonstrated its effectiveness in speech command recognition. However, few studies have conducted extensive research on leveraging attention mechanisms to enhance its performance. In this study, we aim to investigate the deep learning architectures for improved speaker-independent speech command recognition. Specifically, we first compare the log-Mel-spectrogram and log-Gammatone spectrogram using VGG style and VGG-skip style networks. Next, the best-performing model is selected and investigated using different attention mechanisms including channel-time attention, channel-frequency attention, and channel-time-frequency attention. Finally, a dual CNN with cross-attention is used for speech command classification. A self-made dataset including 40 participants with 12 classes is used for the experiment which are all recorded in Mandarin Chinese, utilizing a variety of smartphone devices across diverse settings. Experimental results indicate that using log-Gammatone spectrogram and VGG-skip style networks with cross attention can achieve the best performance, where the accuracy, precision, recall and F1-score are 94.59%, 95.84%, 94.64%, and 94.57%, respectively. © The Author(s), under exclusive licence to Springer Science+Business Media, LLC, part of Springer Nature 2024.

关键词： Spectrographs

来源：评论

学校读者我要写书评

暂无评论

An Optimal Framework for SDN Based on Deep Neural Network

引用

computers, Materials & Continua 2022年第10期73卷 1125-1140页

作者： Abdallah Abdallah Mohamad Khairi Ishak Nor Samsiah Sani Imran Khan Fahad RAlbogamy Hirofumi Amano Samih M.Mostafa Department of Industrial Engineering School of Applied Technical Sciences German Jordanian UniversityAmman35247Jordan School of Electrical and Electronic Engineering Universiti Sains MalaysiaNibong Tebal14300Malaysia Center for Artificial Intelligence Technology Faculty of Information Science and TechnologyThe National University of Malaysia(UKM)Bangi43600SelangorMalaysia Department of Electrical Engineering University of Engineering and TechnologyPeshawar814Pakistan Turabah University College Computer Sciences ProgramTaif UniversityTaif21944Saudi Arabia Research Institute for Information Technology Kyushu UniversityFukuoka819-0395Japan Computer Science-Mathematics Department Faculty of ScienceSouth Valley UniversityQena83523Egypt

Software-defined networking(SDN)is a new paradigm that promises to change by breaking vertical integration,decoupling network control logic from the underlying routers and switches,promoting(logical)network control centralization,and introducing network ***,the controller is similarly vulnerable to a“single point of failure”,an attacker can execute a distributed denial of service(DDoS)attack that invalidates the controller and compromises the network security in *** address the problem of DDoS traffic detection in SDN,a novel detection approach based on information entropy and deep neural network(DNN)is *** approach contains a DNN-based DDoS traffic detection module and an information-based entropy initial inspection *** initial inspection module detects the suspicious network traffic by computing the information entropy value of the data packet’s source and destination Internet Protocol(IP)addresses,and then identifies it using the DDoS detection module based on *** assaults were found when suspected irregular traffic was *** reveal that the algorithm recognizes DDoS activity at a rate of more than 99%,with a much better accuracy *** false alarm rate(FAR)is much lower than that of the information entropy-based detection ***,the proposed framework can shorten the detection time and improve the resource utilization efficiency.

关键词： Deep neural network computer networks data security optimization

来源：评论

学校读者我要写书评

暂无评论

Data Transfer for YOLOv5 Based on Selector Network and Partial Pseudo-Labels

Data Transfer for YOLOv5 Based on Selector Network and Parti...

引用

Cognitive Computing and Complex Data (ICCD), International Conference on the

作者： Guoqiang Qu Jianchu Lin Haiyi Bian Haiping Zhou Muhammad Raees UL Haq Bo Jiang Faculty of Computer and Software Engineering Huaiyin Institute of Technology Huai’an China Faculty of Electronic Information Engineering Huaiyin Institute of Technology Huai’an China Jiangsu Eazytec Co. LTD Yixin China

ISBN: (数字)9798350352894

ISBN: (纸本)9798350352900

YOLOv5 is a widely used object detection method for its high-performance of balancing speed and accuracy. Data transfer for YOLOv5 is a challenging task, as it could involve domain shift between different datasets. We propose a data transfer method for YOLOv5 based on a Selector Network and Partial Pseudo-Labeling. The backbone network of YOLOv5 is used for feature extraction, a Selector Network is employed to filter source domain outlier samples to address the domain shift issue, and Partial Pseudo-Labels are leveraged to improve classification performance in the target domain. The results show that our method improves detection accuracy by an average of 2.4% on public datasets (COCO and VOC) and our datasets, with a notable improvement of 3.1% under complex backgrounds and varying lighting conditions. It demonstrates that our method can effectively enhances the accuracy and stability of YOLOv5.

关键词： YOLO Training Accuracy Filtering Lighting Data transfer Feature extraction Cognitive systems Faces Batch normalization

来源：评论

学校读者我要写书评

暂无评论

Application of Artificial Intelligence using Mammograms to Identify Breast Cancer

Application of Artificial Intelligence using Mammograms to I...

引用

International Conference on Computational Intelligence and Networks (CINE)

作者： Sudeshna Rath Kalyanbrata Giri Satya Ranjan Dash Paola Barra Azian Azamimi Abdullah School of Computer Engineering KIIT University India Department of Science and Technology University of Naples “Parthenope“ Italy Faculty of Electronic Engineering and Technology Universiti Malaysia Perlis Malaysia

ISBN: (数字)9798331516796

ISBN: (纸本)9798331516802

An affordable and precise way to identify breast cancer is through Mammography. It has significantly decreased mortality by detecting malignancies early and is essential for early detection. It has also increased survival rates. However, disparities in mammography access are still mostly driven by socioeconomic factors. Ongoing research aims to address these problems, improve screening methods, and raise the precision and usability of mammography. Additionally, radiologists are being assisted by computer-aided detection (CAD) and artificial intelligence (AI), which are being used to improve diagnostic accuracy. The goal of this research is to identify the best AI algorithms for breast cancer detection. We employ a range of preprocessing methods in this study, first resizing the images to see which method cleans up our data the best. After segmenting our images to locate the region of interest, we extract features using a local binary pattern. Numerous classification models have been employed, such as SVM, KNN, Random Forest, and Decision Tree. Moreover, CNN and RCN have been used for the same objective. Furthermore, it was shown that, generally speaking, neural networks outperformed classification.

关键词： Support vector machines Solid modeling Accuracy Filtering Neural networks Feature extraction Breast cancer Classification algorithms Artificial intelligence Random forests

来源：评论

学校读者我要写书评

暂无评论

A Review on the Advancement of Renewable Natural Fiber Hybrid Composites: Prospects, Challenges, and Industrial Applications

引用

Journal of Renewable Materials 2024年第7期12卷 1237-1290页

作者： Mohammed Mohammed Jawad K.Oleiwi Aeshah M.Mohammed Anwar Ja’afar Mohamad Jawad Azlin F.Osman Tijjani Adam Bashir O.Betar Subash C.B.Gopinath Center of Excellence Geopolymer&Green Technology(CEGeoGTech) Universiti Malaysia PerlisArau02600Malaysia Faculty of Chemical Engineering Technology Universiti Malaysia Perlis(UniMAP)Arau02600Malaysia Department of Materials Engineering University of TechnologyBaghdad10070Iraq Department of Chemical Materials University of Bagdad College of Education for Pure Science Ibn-AlhaithamBaghdad10001Iraq Department of Computer Techniques Engineering Al-Rafidain University CollegeBaghdad10001Iraq Faculty of Electronics Engineering Technology Universiti Malaysia PerlisKampus Uniciti Alam Sg.ChuchuhPerlis02100Malaysia Research Center(NANOCAT) University of MalayaKuala Lumpur50603Malaysia Institute of Nano Electronic Engineering Universiti Malaysia PerlisPerlis01000Malaysia Center for Global Health Research Saveetha Medical College&HospitalSaveetha Institute of Medical and Technical Sciences(SIMATS)Tamil Nadu602105India

Naturalﬁbre(NFR)reinforced functional polymer composites are quickly becoming an indispensable sustainable material in the transportation industry because of their lightweight,lower cost in manufacture,and adaptability to a wide variety of ***,the major difﬁculties of using theseﬁbres are their existing poor dimensional stability and the extreme *** assessing the mechanical properties(MP)of composites,the interfacial bonding(IB)happening between the NFR and the polymer matrix(PM)plays an incredibly signiﬁcant *** compared to NFR/syntheticﬁbre hybrid composites,hybrid composites(HC)made up of two separate NFR are less prevalent;yet,these hybrid composites also have the potential to be valuable materials in terms of environmental issues.A new dimension to theﬂexibility of composites reinforced with NFR is added by the cost-effective manufacture of hybrid composites utilising *** purpose of this study is to offer an over-view of the keyﬁndings that were presented on hybrid *** emphasis was focused on the factors that inﬂuence the performance of the naturalﬁber composites,diverse approaches to enhancing MP,physical,electri-cal,and thermal characteristics of the *** study in polymer science gains interest for applications in con-struction and automotive industries.

关键词： Renewable naturalﬁber hybrid composites sustainability nanomaterial greener material

来源：评论

学校读者我要写书评

暂无评论

The Electrical Properties of a-IGZO TFT Prediction Based on Variational Autoencoder Model

The Electrical Properties of a-IGZO TFT Prediction Based on ...

引用

2023 International Conference on Consumer electronics - Taiwan, ICCE-Taiwan 2023

作者： Hu, Shih-Shin Bea, Kean-Thye Lin, Wei-Hsuan Lin, Da-Zheng Chen, Xiu-Zhi Lin, Ting-Ru Hu, Hsin-Hui Chen, Yen-Lin Chen, Kun-Ming Cheng, Wai Khuen National Taipei University of Technology Department of Computer Science and Information Engineering Taipei106 Taiwan National Taipei University of Technology Department of Electronic Engineering Taipei106 Taiwan Taiwan Semiconductor Research Institute Hsinchu30078 Taiwan Universiti Tunku Abdul Rahman Faculty of Information and Communication Technology Kampar31900 Malaysia

ISBN: (纸本)9798350324174

In this study, we integrated the Polynomial Regression module with a Variational Autoencoder (PolyReg-VAE) model to foster the inference of transfer curves before further deriving the electrical properties of a-IGZO TFT. This approach is able to predict the electrical characteristics of a-IGZO-TFT with different process parameters in a short time, reducing the cost and the time of developing new technology. © 2023 IEEE.

关键词： Deep learning

来源：评论

学校读者我要写书评

暂无评论

Efficient Frameworks for EEG Epileptic Seizure Detection and Prediction

引用

Annals of Data Science 2022年第2期9卷 393-428页

作者： Emara, Heba M. Elwekeil, Mohamed Taha, Taha E. El-Fishawy, Adel S. El-Rabaie, El-Sayed M. El-Shafai, Walid El Banby, Ghada M. Alotaiby, Turky Alshebeili, Saleh A. Abd El-Samie, Fathi E. Department of Electronics and Electrical Communications Engineering Faculty of Electronic Engineering Menoufia University Menouf32952 Egypt Security Engineering Laboratory Department of Computer Science Prince Sultan University Riyadh11586 Saudi Arabia Department of Industrial Electronics and Control Engineering Faculty of Electronic Engineering Menoufia University Menouf32952 Egypt KACST Riyadh Saudi Arabia King Saud University Riyadh Saudi Arabia Department of Electrical Engineering King Saud University Riyadh11421 Saudi Arabia Department of Information Technology College of Computer and Information Sciences Princess Nourah Bint Abdulrahman University Riyadh21974 Saudi Arabia

Seizure detection and prediction are a very hot topics in medical signal processing due to their importance in automatic medical diagnosis. This paper presents three efficient frameworks for applications related to electroencephalogram (EEG) signal processing. The first one is an automatic seizure detection framework based on the utilization of scale-invariant feature transform (SIFT) as an extraction tool. The second one depends on the utilization of the fast Fourier transform (FFT) and an artificial neural network for epileptic seizure prediction. Finally, an automated patient-specific framework for channel selection and seizure prediction is presented based on FFT. The simulation results show the success of the proposed frameworks for automated medical diagnosis. © 2021, Springer-Verlag GmbH Germany, part of Springer Nature.

关键词： Fast Fourier transforms

来源：评论

学校读者我要写书评

暂无评论

FGRL-Net: Fine-Grained Personalized Patient Representation Learning for Clinical Risk Prediction Based on EHRs

FGRL-Net: Fine-Grained Personalized Patient Representation L...

引用

2023 IEEE International Conference on Systems, Man, and Cybernetics, SMC 2023

作者： Chio, KaKit Zhu, Wenhao He, Lihua Zhang, Dian Yang, Xu Luo, Wuman Macao Polytechnic University Faculty of Applied Sciences S.A.R Macao China School of Information and Communication Engineering University of Electronic Science and Technology of China Sichuan Province China Shenzhen University Department of Computer Science and Software Engineering Shenzhen China

ISBN: (纸本)9798350337020

Personalized patient representation learning (PPRL) is a critical element in clinical risk prediction. It aims to obtain a complete portrait of each patient based on electronic Health Records (EHR). Although existing works have achieved remarkable progress in healthcare prediction, there are still three major issues. First, feature correlation is crucial for risk prediction, but it has not yet been fully exploited by existing works. Second, variation pattern of dynamic feature contains useful information about patient's physical status, but adaptive pattern recognition is still a challenge. Third, existing works usually adopt a two-stage embedding process to process each dimension of the EHR data. However, some useful low-level information for PPRL will be lost. To address these issues, in this paper, we propose a fine-grained PPRL architecture named FG RL- N et for clinical risk prediction based on EHR. Specifically, we propose a Medical Feature Correlation Detection Module (FCM) to effectively learn the feature correlations for each patient and a Temporal Variation Pattern Recognition Module (TVM) to effectively detect the variation patterns of each dynamic feature. Moreover, we design a Fine-Grained Representation Mechanism (FGRM) to preserve the low-level information (from both feature and visit dimensions) useful for risk prediction. In addition, in the stage of data preprocessing, We utilize generic medical classification knowledge to classify numerical dynamic data. We conduct the in-hospital mortality experiment and the decompensation experiment on a real-world dataset. The experiment results show that the FGRL-Net outperforms state-of-the-art approaches. The source code is provided in github https://***/JackyChio/FGRL-Net. © 2023 IEEE.

关键词：

来源：评论

学校读者我要写书评

暂无评论

Resource Assessment of Wind Energy Potential of Mokha in Yemen with Weibull Speed

引用

computers, Materials & Continua 2021年第10期69卷 1123-1140页

作者： Abdulbaset El-Bshah Fahd N.Al-Wesabi Ameen M.Al-Kustoban Mohammad Alamgeer Nadhem Nemri Majdy M.Eltahir Hany Mahgoub Noha Negm Department of Electronic Engineering Faculty of EngineeringUniversity of Science and Technology UST-Sana’aYemen Department of Computer Science King Khalid UniversityMuhayel AseerKingdom of Saudi Arabia Faculty of Computer and IT Sana’a UniversitySana’aYemen Department of Electronic Faculty of EngineeringUSTSana’aYemen Department of Information Systems King Khalid UniversityMayahel AseerKSA Department of Computer Science King Khaled UniversityKSA&Faculty of Computers and InformationMenoufia UniversityEgypt Department of Computer Science King Khaled UniversityKSA&Faculty of ScienceDepartment of Mathematics and Computer ScienceMenoufia UniversityEgypt

The increasing use of fossil fuels has a significant impact on the environment and ecosystem,which increases the rate of *** the high potential of renewable energy sources in Yemen and other Arabic countries,and the absence of similar studies in the *** study aims to examine the potential of wind energy in Mokha *** was done by analyzing and evaluating wind properties,determining available energy density,calculating wind energy extracted at different altitudes,and then computing the capacity factor for a few wind turbines and determining the *** speed was verified as the closest to the average actual wind speed using the cube root,as this was verified using 3 criteria for performance analysis methods(R^(2)=0.9984,RMSE=0.0632,COE=1.028).The wind rose scheme was used to determine the appropriate direction for directing the wind turbines,the southerly direction was appropriate,as the winds blow from this direction for 227 days per year,and the average southerly wind velocity is 5.27 m/s at an altitude of 3 *** turbine selected in this study has a tower height of 100m and a rated power of 3.45 *** capacitance factor was calculated for the three classes of wind turbines classified by the International Electrotechnical Commission(IEC)and compared,and the turbine of the first class was approved,and it is suitable for the study site,as it resists storms more than *** daily and annual capacity of a single,first-class turbine has been assessed to meet the needs of 1,447 housing units in Mokha *** amount of energy that could be supplied to each dwelling was around 19 kWh per day,which was adequate to power the basic loads in the home.

关键词： Wind energy system probability density wind speed Weibull velocity Rayleigh velocity wind rose

来源：评论

学校读者我要写书评

暂无评论

MySkills - A Reflective Framework for Employability Skills 14

MySkills - A Reflective Framework for Employability Skills

引用

14th IEEE Global engineering Education Conference, EDUCON 2023

作者： Naeem, Usman Bosman, Lisa Revell, Claire School of Electronic Engineering and Computer Science Queen Mary University of London London United Kingdom Purdue University Department of Technology Leadership and Innovation West Lafayette United States Queen Mary University of London Faculty of Science and Engineering London United Kingdom

ISBN: (纸本)9798350399431

Learner retention and employability are important metrics within the higher education sector, which places an onus on higher education institutions to ensure learners not only complete their degree programmes but also get employed in graduate-level roles after graduation. Given the competency nature of computer Science (CS) and electronic engineering (EE) programmes, it is easier for these CS/EE learners (in comparison to other majors throughout the university) to demonstrate the skills required by employers within the digital sector. However, a challenge exists in that it can be difficult for learners to see the connections between the core skills taught in their modules and occupational areas. The work in this paper describes 'mySkills', which is a reflective framework that allows learners to track and reflect on their skills-level progression throughout their degree programme. This framework was introduced in the academic year 2021/22 as a pilot within the School of electronic engineering and computer Science. © 2023 IEEE.

关键词： E-learning

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：