检索结果-内蒙古大学图书馆

An Evaluation of Deep Learning Models for Classifying Time Series Individual Data Instances

Journal of computer and Communications 2024年第11期12卷 187-206页

作者： Joshua A. Blaney Suresh S. Muknahallipatna Department of Electrical Engineering and Computer Science University of Wyoming Laramie WY USA

Deep learning for time series sequence individual data instance classification can revolutionize computer assisted navigation by providing surgeons with accurate, real-time instrument locality through automatic instrument localization. This paper presents an evaluation of Deep Learning models to perform individual data instance classification of time series data. The models explored include convolution and recurrent networks, as well as state-of-the-art residual and inception architectures. The time series data used to evaluate the models consists of depth and force measurements from a drill. Four recurrent neural network models using long short-term memory and gated recurrent units, known as baseline models, and four models using 1D convolution with ResNet and Inception architectures, known as advanced models, were evaluated by determining the data instance membership of the four classes. The four classes represent four distinct regions in a bone traversed by the drill bit during a surgical procedure. First, the time series data is preprocessed, identifying the four classes or regions of the bone. Next, the paper presents a discussion of the network architecture and modifications of both the basic and advanced deep learning models, followed by the training process and hyperparameters tuning. The performance of the models was evaluated using the precision and recall performance parameters. Out of the eight models evaluated, the recurrent neural network with gated recurrent units has the best performance. The paper also demonstrates the importance of the feature depth over the feature force in classifying the data instances, followed by the effects of the imbalanced dataset on the performance of the models.

关键词： Recurrent Neural Networks Long Short Term Memory Gated Recurrent Units Residual Networks Inception Networks

来源：评论

学校读者我要写书评

暂无评论

Edge Computing for Efficient Storage and Low-Latency Video Streaming in Cloud Environments

Edge Computing for Efficient Storage and Low-Latency Video S...

引用

2023 IEEE International Conference on Artificial Intelligence, Blockchain, and Internet of Things, AIBThings 2023

作者： Darwich, Mahmoud Khalil, Kasem Ismail, Yasser Bayoumi, Magdy University of Mount Union Department of Mathematics and Computer Science AllianceOH44601 United States University of Mississippi University Department of Electrical and Computer Engineering MS38677 United States Southern University and AandM College Electrical Engineering Department Baton RougeLA70807 United States University of Louisiana at Lafayette Department of Electrical and Computer Engineering LA70504 United States

ISBN: (纸本)9798350322347

Efficient storage and low-latency video streaming are critical for delivering high-quality multimedia experiences in cloud environments. This research explores the potential of edge computing as a solution to address these challenges. By strategically deploying computational resources at the network edge, edge computing minimizes latency and enhances storage capabilities, leading to an improved user experience. We propose a cutting-edge prototype system that integrates edge computing into the cloud infrastructure. Through rigorous experiments, we evaluate key performance metrics, including latency, bitrate, and storage utilization, the results show the effectiveness of the proposed approach for video streaming in the cloud. Particularly, it reduces the latency by up to 25% and storage utilization by up to 15%. © 2023 IEEE.

关键词： Edge computing

来源：评论

学校读者我要写书评

暂无评论

Telemedical Robot Using IoT with Live Supervision and Emergency Alert 3

Telemedical Robot Using IoT with Live Supervision and Emerge...

引用

3rd International Conference on Pervasive Computing and Social Networking, ICPCSN 2023

作者： Vimala, S. Karthika, G. Geetha, C. Reddy, R. Vijaya Kumar Ganesh, C. S. Sundar Ganesh, R. Karthik Panimalar Engineering College Department of Electrical and Electronics Engineering Tamil Nadu Chennai600123 India R.M.D. Engineering College Department of Electronics and Communication Engineering Tamil Nadu Kavaraipettai601206 India R. M. K. Engineering College Department of Computer Science and Engineering Tamil Nadu Kavaraipettai601206 India Koneru Lakshmaiah Education Foundation Department of Computer Science and Engineering Andhra Pradesh Vaddeswaram522502 India Karpagam College of Engineering Department of Electrical and Electronics Engineering Tamil Nadu Coimbatore641032 India Scad College of Engineering and Technology Departrtent of Computer Science and Engineering Tamil Nadu Tirunelveli627414 India

ISBN: (纸本)9798350322842

A highly qualitative service is required for monitoring and assisting the remote patients as the telemedicine industry soars to new heights. When patients are far, doctors find it challenging to interact with them and diagnose their health conditions. In addition, individuals find it challenging to administer injections on their own in an emergency. This project provides a design for a medical assistance, real-time monitoring, and injection robot as well as human-robot interaction (HRI) in the field of healthcare for a robot known as a Telemedicine robot that runs on an IOT platform. Based on a live video conference and saved in an electronic medical record, it can be utilized for remote diagnostics and injecting patients. The Arduino platform is used for IOT functionality wherein the doctor can command the robot through mobile for locomotion. When necessary, the doctor in the range can deploy a set of sensors to keep an eye on the patient's wellbeing and store the information of the patient in the database and control the robot to put injection. The project emphasizes low cost and modular design with customizable sensor modules. The IOT capabilities of the Arduino platform allow the doctor to control the robots movement using a mobile device. When necessary, the doctor from a distance can deploy a collection of sensors to keep an eye on the patients wellbeing, save the patients data in a database, and command a robot to administer an injection. The project places a strong emphasis on modularity, cheap cost, and adaptable sensor modules. © 2023 IEEE.

关键词： Telemedicine

来源：评论

学校读者我要写书评

暂无评论

Advancing Long-Wavelength InGaN Micro-LEDs as Wavelength-Selective Detectors for White-Light Communication

引用

ACS Photonics 2024年第8期11卷 3390-3400页

作者： Lu, Tingwei Dai, Yurong Lee, Tzu-Yi Zheng, Xi Chen, Guolong Su, Yuhan Chen, Zhong Kuo, Hao-Chung Wu, Tingzhu Fujian Engineering Research Center for Solid-State Lighting Department of Electronic Science Xiamen University Xiamen361005 China National Integrated Circuit Industry and Education Integration Innovation Platform Department of Electronic Science Xiamen University Xiamen361005 China Department of Photonics Graduate Institute of Electro-Optical Engineering College of Electrical and Computer Engineering National Yang Ming Chiao Tung University Hsinchu30010 Taiwan

Long-wavelength InGaN-based microlight emitting diodes (μLEDs) have the potential for novel display and visible light communication (VLC). In this study, red and yellow μLEDs were fabricated for VLC applications as VLC transmitters and photodiodes (PDs). Stress modulation engineering and atomic layer deposition (ALD) techniques were employed to improve the transmission and detection performance. When operating as VLC transmitters, the red and yellow μLEDs exhibits high modulation bandwidth of 439.7 and 532.5 MHz at 2000 A/cm2, achieving a data rate of 1.9 and 2.4 Gbps, respectively. When operating as VLC PDs, long-wavelength emitting μLEDs with high indium component QW exhibit longer wavelength absorption edge, which is expected to achieve higher responsivity to blue light while filtering out the fluorescence component. The response spectra of the red and yellow μLED PDs overlap a large portion of the blue signal. The red and yellow μLEDs achieved responsivities of up to 0.208 and 0.135 A/W, respectively, for 450 nm light, higher than those of the reported InGaN LED-based photodetectors. In addition, owing to the wavelength selectivity, the long-wavelength μLED PDs could filter the slow fluorescence emission, thus achieving white light modulation bandwidth several times that achieved by adopting a silicon-based detector. A white-light VLC system was experimentally demonstrated. Maximum transmission rates of 1.2 and 1.05 Gbps were achieved based on red and yellow μLEDs operating as PDs, respectively. The findings of this study indicate that employing red μLED pixels as wavelength-selective PDs in future μLED information displays is an effective strategy for phosphor-based white-light communication. © 2024 American Chemical Society

关键词： Photodetectors

来源：评论

学校读者我要写书评

暂无评论

Fine-Grained Global Modeling Learning for Personalized Federated Sequential Recommender

Fine-Grained Global Modeling Learning for Personalized Feder...

引用

2025 IEEE International Conference on Acoustics, Speech, and Signal Processing, ICASSP 2025

作者： Di, Yicheng Shi, Hongjian Ma, Ruhui Liu, Yuan Wang, Weiyu School of Artificial Intelligence and Computer Science Jiangnan University Wuxi China School of Electronic Information and Electrical Engineering Shanghai Jiao Tong University Shanghai China Department of Computer Engineering San Jose State University San Jose United States

ISBN: (纸本)9798350368741

Personalized sequential recommender has become a key task in the consumer electronics domain. Existing methods for personalized sequential recommenders primarily focus on modeling user behavior and have achieved satisfactory recommender results. However, the inherent quadratic computational complexity in most existing methods often causes typically results in inefficiencies, impeding real-time suggestions. Additionally, these methods cannot be fine-tuned to the personalized needs of users across different scenarios. To address these challenges, we propose the Fine-Grained Global Modeling Learning for Personalized Federated Sequential Recommender (FedSR). Specifically, we design the Associative Mamba Block to model user profiles from a global perspective and enhance prediction efficiency. Furthermore, we introduce the Variable Response Mechanism to fine-tune parameters according to the personalized needs of users. Additionally, we design the Dynamic Magnitude Loss to retain more local personalized information during training. Extensive experiments on three real-world datasets confirms that the proposed FedSR surpasses current approaches in terms of both performance and efficiency, up to 9.48% improvement. © 2025 IEEE.

关键词： federated learning federated recommender mamba sequential recommender

来源：评论

学校读者我要写书评

暂无评论

Hybrid Local Cache Model for Enhancing Performance and Memory Footprint of Mobile Heap Allocator 12

Hybrid Local Cache Model for Enhancing Performance and Memor...

引用

12th IEEE Global Conference on Consumer Electronics, GCCE 2023

作者： Lee, Kyung-Seok Lee, Teahyung Eom, Young Ik Sungkyunkwan University Department of Digital Media and Communications Engineering Suwon Korea Republic of Sungkyunkwan University Department of Computer Science and Engineering Suwon Korea Republic of Sungkyunkwan University College of Computing and Informatics Department of Electrical and Computer Engineering Suwon Korea Republic of

ISBN: (纸本)9798350340181

The representative mobile heap allocator that has been used in Android system, called SCUDO, supports two local cache models: the shared model and the exclusive model. However, neither the shared model nor the exclusive model cannot satisfy both memory usage and performance requirements simultaneously since mobile applications have diverse memory allocation/free patterns and the number of threads varies from single to several tens. This paper suggests a hybrid local cache model that aims to adaptively meet memory usage and performance goals. Our evaluation results show that our hybrid local cache model outperforms the shared model by up to 22% on allocation time, while preserving the memory usage. © 2023 IEEE.

关键词： Android (operating system)

来源：评论

学校读者我要写书评

暂无评论

Transformer Monitoring and Security System Using IoT 3

Transformer Monitoring and Security System Using IoT

引用

3rd International Conference on Innovative Practices in Technology and Management, ICIPTM 2023

作者： Anakal, Sudhir Kar, Sangita Rani Sangeetha, A. Sritha, P. Satish, Dekka Sajitha, P.L. Prasad, A. Rajendra Sharnbasva University Department of Computer Applications Karnataka India ITER Siksha 'o' Anusandhan Department of Electrical & Electronics Odisha India R.M.K College of Engineering & Technology Department of Computer Science Chennai India Bannari Amman Institute of Technology Department of Electrical & Electronics Chennai India Lendi Institute of Engineering & Technology Department of Computer Science Andhra Pradesh India R. M. K Engineering College Department of Computer Science Chennai India Sri Sai Ram Engineering College Department of Mechanical Engineering Chennai India

ISBN: (纸本)9798350336238

Temperature increases are a common source of problems in transformers. Temperature increases as load current increases. There are two ways to monitor the rise in load current. Use a potential transformer and a temperature sensor, respectively. Voltage drops and winding temperature rises as load current increases. By measuring these two variables with a voltage sensor and a temperature sensor, the transformer problem can be resolved. The quantity detected by the sensor will be relayed over IoT to the control room in order to inform it of the transformer's state. It operates totally on autopilot. The Node MCU8266 board (which consist of microcontroller and built-in Wi-Fi board) included with the sensor will activate the circuit breaker to protect the transformer when the fault level is high. © 2023 IEEE.

关键词： Temperature sensors

来源：评论

学校读者我要写书评

暂无评论

Cyberbullying Recognition on social media for an Afro-Asiatic language Content Using Attention Mechanism and Bi-RNN Classifier

引用

Procedia computer science 2025年 258卷 2232-2243页

作者： Rajesh Sharma R Muluken Ayele Alemu Nune Sreenivas Akey Sungheetha Sheila Mahapatra Department of Computer Science and Engineering Alliance University Bangalore-India Department of Computer Science College of Business and Economics Arsi University Asella Ethiopia Department of Computer Science and Engineering Adama Science and Technology University Adama Ethiopia Department of Electrical and Electronic Engineering Alliance University Bangalore-India

In the present era, cyberbullying on social media has developed into a complicated issue. In Ethiopia, cyberbullying involving sexual content material has become a prevalent issue in recent times. Because the content material on social media is unstructured, cyberbullying that is just motivated by sexual texts on the platform can be a laborious and complicated process. The difficulty of finding sexual texts on public media has raised in recent years, leading some experts to focus their attention on the detection of cyberbullying. Because deep learning algorithms do greatly on tasks involving natural language processing, several academics have suggested using these models to detect cyberbullying. Additionally, the majority of research on this challenge count number has focused on socio-political context, handicap, religion, and ethnicity. As a remedy, this observer suggested a thorough investigation of cyberbullying detection for Amharic sexual writing on social media. To create models, a binary Amharic sexual dataset is paired with a sexual dataset trained from collected Amharic content from the Facebook platform. "Bullying" and "Non-bullying" are the binary classes that make up the prepared dataset. Because the W2Vec model works well for describing non-unusual place key phrases in constrained period datasets, it is mostly based on a Skip-gram model. Furthermore, GRU and LSTM networks are used for model comparison, while the Bi-RNN and Attention Mechanism models are employed for classification. Trends in the application of N-fold cross-validation are understood. N-fold cross-validation on our dataset yields very good universal overall performance, according to the results.

关键词： Cyberbullying detection Deep Learning Natural language processing Amharic language sexual content Bidirectional RNN

来源：评论

学校读者我要写书评

暂无评论

NDN Collaborative Caching Replacement and Placement Policy Performance Evaluation 10

NDN Collaborative Caching Replacement and Placement Policy P...

引用

10th International Conference on Wireless and Telematics, ICWT 2024

作者： Zaki Hamidi, Eki Ahmad Akbar, Khibran Muhammad Negara, Ridha Muldina Payangan, Ivah Puspita Puspitaningsih, Mutiara Dewi Putri As'ari, Adzka Zahratu Jupriyadi UIN Sunan Gunung Djati Bandung Electrical Engineering Department Bandung Indonesia Telkom University School of Electrical Engineering Bandung Indonesia The University Center of Excellence Intelligent Sensing-IoT Telkom University Indonesia Telkom University School of Electrical Engineering Indonesia Institut Teknologi Bandung School of Electrical Engineering and Informatics Bandung Indonesia Universitas Teknokrat Indonesia Faculty of Engineering and Computer Science Bandarlampung Indonesia

ISBN: (纸本)9798350377057

Named Data Networking (NDN) is an emerging technology that aims to provide rapid and efficient content distribution and retrieval in the network. This paper focuses on the impact of selecting the pair between cache replacement and placement within NDN, an aspect often overlooked in previous research. Four cache placement strategies will be implemented in the initial simulation: Leave Copy Down (LCD), Leave Copy Everywhere (LCE), Prob Cache, and Cache Less for More (CL4M). Each cache placement will be combined with two replacements: least frequently used (LFU) and least recently used (LRU). The effects of these combinations on critical metrics such as cache hit ratio, latency, link load, and path stretch are investigated. The results reveal that LCE combined with LFU demonstrates remarkable enhancements in all measurements over LCE coupled with LRU. These findings underscore the importance of meticulous selection and optimization of cache placement and replacement strategies in NDN environments to augment network efficiency and performance. © 2024 IEEE.

关键词： Neodymium alloys

来源：评论

学校读者我要写书评

暂无评论

Condition Monitoring and Maintenance Management with Grid-Connected Renewable Energy Systems

引用

computers, materials & Continua 2022年第8期72卷 3999-4017页

作者： Md.Mottahir Alam Ahteshamul Haque Mohammed Ali Khan Nebras M.Sobahi Ibrahim Mustafa Mehedi Asif Irshad Khan Department of Electrical and Computer Engineering(ECE) King Abdulaziz UniversityJeddah 21589Saudi Arabia Advance Power Electronics Research Lab Department of Electrical EngineeringJamia Millia IslamiaNew Delhi 110025India Department of Electrical Power Engineering Faculty of Electrical Engineering and CommunicationBrno University of TechnologyBrno 61600Czech Republic Center of Excellence in Intelligent Engineering Systems(CEIES) King Abdulaziz UniversityJeddah 21589Saudi Arabia Computer Science Department Faculty of Computing and Information TechnologyKing Abdulaziz UniversityJeddah 21589Saudi Arabia

The shift towards the renewable energy market for carbon-neutral power generation has encouraged different governments to come up with a plan of *** with the endorsement of renewable energy for harsh environmental conditions like sand dust and snow,monitoring and maintenance are a few of the prime *** problems were addressed widely in the literature,but most of the research has drawbacks due to long detection time,and high misclassification *** to overcome these drawbacks,and to develop an accurate monitoring approach,this paper is motivated toward the understanding of primary failure concerning a grid-connected photovoltaic(PV)system and highlighted along with a brief overview on existing fault detection *** on the drawback a data-driven machine learning approach has been used for the identification of fault and indicating the maintenance unit regarding the operation and maintenance ***,the system was tested with a 4 kWp grid-connected PV system,and a decision tree-based algorithm was developed for the identification of a *** results identified 94.7%training accuracy and 14000 observations/sec prediction speed for the trained classifier and improved the reliability of fault detection nature of the grid-connected PV operation.

关键词： Fault detection machine learning fault ride through

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：