检索结果-内蒙古大学图书馆

arXiv 2024年

作者： Yi, Jingyuan Yu, Peiyang Huang, Tianyi Xu, Zeqiu Information Networking Institute Carnegie Mellon University PittsburghPA15213 United States Department of Electrical Engineering & Computer Sciences University of California Berkeley BerkeleyCA94720 United States

Aiming at the latest particle swarm optimization algorithm, this paper proposes an improved Transformer model to improve the accuracy of heart disease prediction and provide a new algorithm idea based on particle swarm optimization (PSO). We first use three mainstream machine learning classification algorithms - decision tree, random forest and XGBoost, and then output the confusion matrix of these three models. The results showed that the random forest model had the best performance in predicting the classification of heart disease, with an accuracy of 92.2%. Then, we apply the Transformer model based on PSO algorithm to the same dataset for the classification experiment. The results show that the classification accuracy of the model is as high as 96.5%, 4.3% higher than that of random forest, which verifies the effectiveness of PSO in optimizing Transformer model. The above research shows that PSO significantly improves Transformer performance in heart disease prediction. Improving the ability to predict heart disease is a global priority with benefits for all humankind. Accurate prediction can enhance public health, optimize medical resources, and reduce healthcare costs, leading to a healthier society. This advancement paves the way for more efficient health management and supports the foundation of a healthier, more resilient global community. © 2024, CC BY.

关键词： Prediction models

来源：评论

学校读者我要写书评

暂无评论

BitMoD: Bit-serial Mixture-of-Datatype LLM Acceleration

arXiv

引用

arXiv 2024年

作者： Chen, Yuzong AbouElhamayed, Ahmed F. Dai, Xilai Wang, Yang Andronic, Marta Constantinides, George A. Abdelfattah, Mohamed S. Computer Systems Lab Cornell University United States Systems and Networking Research Group Microsoft Research United States Department of Electrical and Electronic Engineering Imperial College London United Kingdom

Large language models (LLMs) have demonstrated remarkable performance across various machine learning tasks. Yet the substantial memory footprint of LLMs significantly hinders their deployment. In this paper, we improve the accessibility of LLMs through BitMoD1, an algorithm-hardware co-design solution that enables efficient LLM acceleration at low weight precision. On the algorithm side, BitMoD introduces fine-grained data type adaptation that uses a different numerical data type to quantize a group of (e.g., 128) weights. Through the careful design of these new data types, BitMoD is able to quantize LLM weights to very low precision (e.g., 4 bits and 3 bits) while maintaining high accuracy. On the hardware side, BitMoD employs a bit-serial processing element to easily support multiple numerical precisions and data types;our hardware design includes two key innovations: First, it employs a unified representation to process different weight data types, thus reducing the hardware cost. Second, it adopts a bit-serial dequantization unit to rescale the per-group partial sum with minimal hardware overhead. Our evaluation on six representative LLMs demonstrates that BitMoD significantly outperforms state-of-the-art LLM quantization and acceleration methods. For discriminative tasks, BitMoD can quantize LLM weights to 4-bit with © 2024, CC BY-NC-ND.

关键词：

来源：评论

学校读者我要写书评

暂无评论

Assessing the Effects of Blackhole Attacks on MANET Reliability and Security

Assessing the Effects of Blackhole Attacks on MANET Reliabil...

引用

Decision Aid Sciences and Application (DASA), International Conference on

作者： Amjed Abbas Ahmed Saif Aamer Fadhil Ali Hamza Najim Khattab M. Ali Alheeti Nurhizam Safie Mohd Satar Aisha Hassan Abdallah Hashim Faculty of Information Science and Technology Center for Cyber Security Universiti Kebangsaan Malaysia Bangi Malaysia Department of Computer Techniques Engineering Imam AlKadhum College (IKC) Baghdad Iraq Department of Computer Technical Engineering Imam Al-Kadhum College (IKC) Baghdad Iraq Computer Networking Systems Department College of Computer and Information Technology University of Anbar Anbar Iraq Faculty of Information Science and Technology Center for Cyber Security Universiti Kebangsaan Malaysia (UKM) Bangi Malaysia Department of Electrical and Computer Engineering Faculty of Engineering International Islamic University Malaysia Selangor Malaysia

ISBN: (数字)9798350369106

ISBN: (纸本)9798350369113

Mobile Ad Hoc Networks (MANETs) are characterized by some important attributes, including infrastructure, mobile, and dynamic nature, which makes them have vast applications in different areas of computer networks. The black hole attack falls among the strongly criticized security threats within the framework of the MANETs. This deceptive behavior captures data packets from a source destined for the malicious node and then declines them when they are received. This paper comprehensively reviews prior studies done to analyze and mitigate black hole attacks in MANETs. Besides this, we will conduct a comparative analysis of the research above projects. The present paper extends prior work by actually performing the black hole assault simulations in a MANETs environment. This is done with the aid of the NS2 simulator and the Ad Hoc On-Demands Distance Vector (AODV) routing protocol. In this study, the black hole attack is analyzed and the consequences of the attack are assessed in terms of the system performance and security. We analyze crucial parameters like end-to-end latency, PDR, packet drop, and throughput to evaluate the consequences of this assault. From our results derived from the analysis of our simulations, we aim to present substantial points to do with the impacts of blackhole attacks on MANETs. In this way, our goal is to participate in the perpetually evolving efforts aimed at enhancing the security and reliability of such dynamic networks.

关键词： computer network reliability Collaboration Throughput Ad hoc networks Vectors Security Reliability Mobile computing AODV Mobile ad hoc networks

来源：评论

学校读者我要写书评

暂无评论

Dynamic Encoding and Decoding of Information for Split Learning in Mobile-Edge Computing: Leveraging Information Bottleneck Theory

Dynamic Encoding and Decoding of Information for Split Learn...

引用

IEEE Conference on Global Communications (GLOBECOM)

作者： Omar Alhussein Moshi Wei Arashmid Akhavain Department of Electrical Engineering and Computer Science Khalifa University Abu Dhabi UAE Advanced Networking Team Huawei Ottawa Research & Development Centre Ottawa Canada Department of Computer Science York University Toronto Canada

Split learning is a privacy-preserving distributed learning paradigm in which an ML model (e.g., a neural network) is split into two parts (i.e., an encoder and a decoder). The encoder shares so-called latent representation, rather than raw data, for model training. In mobile-edge computing, network functions (such as traffic forecasting) can be trained via split learning where an encoder resides in a user equipment (UE) and a decoder resides in the edge network. Based on the data processing inequality and the information bottleneck (IB) theory, we present a new framework and training mechanism to enable a dynamic balancing of the transmission resource consumption with the informativeness of the shared latent representations, which directly impacts the predictive performance. The proposed training mechanism offers an encoder-decoder neural network architecture featuring multiple modes of complexity-relevance tradeoffs, enabling tunable performance. The adaptability can accommodate varying real-time network conditions and application requirements, potentially reducing operational expenditure and enhancing network agility. As a proof of concept, we apply the training mechanism to a millimeter-wave (mmWave)-enabled throughput prediction problem. We also offer new insights and highlight some challenges related to recurrent neural networks from the perspective of the IB theory. Interestingly, we find a compression phenomenon across the temporal domain of the sequential model, in addition to the compression phase that occurs with the number of training epochs.

关键词：

来源：评论

学校读者我要写书评

暂无评论

Sustainable Development of a Prototype of Air Mattress from Re-Used Materials for Pressure Ulcer Prevention

Sustainable Development of a Prototype of Air Mattress from ...

引用

2021 International Conference on Green Energy, Computing and Sustainable Technology, GECOST 2021

作者： Daengsi, Therdpong Muttisan, Bundit Wuttidittachotti, Pongpisit Sirawongphatsara, Patsita Rmutp Sustainable Industrial Management Engineering Bangkok Thailand Kmutnb Data Communication and Networking Bangkok Thailand Rmutto Department of Computer Science Thailand

ISBN: (纸本)9781665438650

This paper presents the development of an air mattress to prevent pressure ulcers in patients who have limited movement. The major part of this mattress is a set of re-used dialysis bags that are a kind of waste from hospitals. This mattress prototype has been developed using Arduino Nano boards, force sensitive resistor (FSR) sensors and diaphragm air compressors. After completion, the first version of the air mattress prototype was evaluated by experienced nursing workers as acceptable with a score of 3.24±0.75, based on 5-point scale, which means further improvements for the next version are required. © 2021 IEEE.

关键词： Diseases

来源：评论

学校读者我要写书评

暂无评论

Alzheimer’s Disease Classification via Pre-trained Convolutional Neural Network Variants Utilizing Multi-level Thresholding based on Kapur’s Entropy

Alzheimer’s Disease Classification via Pre-trained Convolut...

引用

Network, Multimedia and Information Technology (NMITCON) International Conference on

作者： Nivedhitha Mahendran P M Durai Raj Vincent Department of Networking and Communications School of Computing Technologies SRM Institute of Science and Technology Chennai Kattankulathur India School of Computer Science Engineering and Information Systems Vellore Institute of Technology Vellore India

ISBN: (数字)9798350372892

ISBN: (纸本)9798350372908

Neurological diseases are the disorders of brain and the nerves that run throughout the human body. There is a range of symptoms that arise as a result of chemical, structural, or electrical modifications in the regions of brain. One such neurological disease is the Alzheimer’s Disease (AD), is considered as a slow and progressive disease that majorly affects the cognitive abilities of the individual. It is most common forms of dementia without any proper cure. However, early diagnosis and treatment might slow down the symptoms and improve the lifestyle of the individual. The most common techniques to identify the neurological disorders are imaging of the brain regions, namely, CT, MRI, PET, etc. The learning approaches, such as Deep Learning and Machine Learning are identified to be effective in analyzing these images to provide better accuracies in detecting the AD. In this study, four versions of Convolutional Neural Networks (CNN) are used to classify AD and Control cases. Further, the Kapur’s thresholding technique is also implemented to segment the images for improving the classification accuracy of AD. From the results shown in the study, the inference shows that the VGG16 of CNN provides considerable improvement than the other deep learning models before and after the implementation of thresholding.

关键词： Deep learning Neurological diseases Image segmentation Accuracy Thresholding (Imaging) Magnetic resonance imaging Computed tomography Convolutional neural networks Alzheimer's disease Information technology

来源：评论

学校读者我要写书评

暂无评论

Smart Water-Powered Renewable Energy System with IoT Integration

Smart Water-Powered Renewable Energy System with IoT Integra...

引用

Decision Aid Sciences and Application (DASA), International Conference on

作者： Amjed Abbas Ahmed Ali Hamza Najim Ahmed Salman Ibraheem Khattab M. Ali Alheeti Nurhizam Safie Mohd Satar Aisha Hassan Abdallah Hashim Center for Cyber Security Faculty of Information Science and Technology Universiti Kebangsaan Malaysia Bangi Malaysia Department of Computer Techniques Engineering Imam AlKadhum College (IKC) Baghdad Iraq Department of Computer Technical Engineering Imam Al-Kadhum College (IKC) Al-Diwaniyah Iraq Department of Computer Technical Engineering Imam Al-Kadhum College (IKC) Baghdad Iraq Computer Networking Systems Department College of Computer and Information Technology University of Anbar Anbar Iraq Center for Cyber Security Faculty of Information Science and Technology Universiti Kebangsaan Malaysia (UKM) Bangi Malaysia Department of Electrical and Computer Engineering Faculty of Engineering International Islamic University Malaysia Selangor Malaysia

ISBN: (数字)9798350369106

ISBN: (纸本)9798350369113

An advanced hybrid renewable energy collecting and storage system prototype that utilizes water flow as its primary power source. Additionally, the system incorporates an energy monitoring system based on the Internet of Things (IoT) technology. The device effectively harnesses energy from the movement of water, storing it for future use in low-voltage applications. It demonstrates a maximum daily energy capture capacity of about 100Wh. The energy monitoring system, based on the Internet of Things (IoT), enables the real-time and remote monitoring of energy harvesting and usage. We thoroughly assessed the energy harvesting component across different situations to identify the crucial parameters that impact energy output. The results of our study demonstrate that several elements, including the spatial separation “between the energy harvester and the water source, Water flow” speed, sensor orientation, and sensor density, all play significant roles in determining the quantity of energy collected. This research shows that it is possible to integrate IoT technology with hydropower renewable energy systems. It provides significant perspectives on enhancing energy extraction efficiency in practical scenarios.

关键词： Renewable energy sources Low voltage Prototypes Real-time systems Internet of Things Energy harvesting Water resources Remote monitoring Sensor arrays Pragmatics

来源：评论

学校读者我要写书评

暂无评论

Intelligent Personal Health Monitoring and Guidance Using Long Short-Term Memory

引用

Journal of Mobile Multimedia 2022年第2期18卷 349-372页

作者： Velliangiri, S. Anbarasu, V. Karthikeyan, P. Anandaraj, S.P. B V Raju Institute of Technology Telangana Narasapur India Department of Networking and Communications School of Computing SRM Institute of Science and Technology Katankulathur Chennai India Computer Science and Engineering Jain (Deemed-to-be University) Bangalore India CSE in Presidency University Bangalore India

Rapid improvements in information technology have made everything in this world contemporary. The mobile phone plays a vital role in the day to day activities. Many mobile applications are developed by using deep learning models to give health guidance to people. We proposed intelligent personal health monitoring and guidance (IPHMG) using long short-term memory to assess the users’ overall health status to solve the mobile application performance problem. The main objective of the research work is to minimize the delay time of the user’s request and improve the accuracy of health predictions. The proposed system calculates scores using the IPHMG score model to find the health conditions of the users. IPHMG score model uses different time-series data to calculate scores such as environment data, body signal data, parent report data, emotion, and health report. Additionally, an Android application is a module that is designed for mobile users to feed their health data and check their health status. The proposed system was implemented. Results show that the proposed method provides better uploading time, processing time, and the user downloading time than simple RNN and ANN methods. © 2021 River Publishers

关键词： Android application Artificial neural network IPHMG LSTM Recurrent neural network

来源：评论

学校读者我要写书评

暂无评论

Smart System for Vechicle Number Plate Recognition Using Convolutional Neural Network(CNN)

Smart System for Vechicle Number Plate Recognition Using Con...

引用

Optimization Techniques for Learning (ICOTL), International Conference on

作者： R Nidhya R Kalpana R Sudhakar M Maranco G Smilarubavathy Department of Computer Science & Engineering Madanapalle Institute of Technology & Science Andhra Pradesh India Department of Networking and Communications SRM Institute of Science and Technology Chennai Tamilandu India Department of Computer Science & Engineering St. Joseph’s Institute of Technology Chennai Tamilandu India

In the era of smart cities and advanced transportation systems, the need for efficient and accurate vehicle identification has become paramount. This paper introduces a sophisticated system for vehicle number plate recognition leveraging the capabilities of Convolutional Neural Networks (CNNs). The proposed system aims to enhance security, streamline traffic management, and contribute to the evolution of intelligent transportation infrastructure. Its is also used to check for suspicious vehicles or vehicles that were involved in crime, in that case it is useful for the polices to browse the details of the accused vehicles. Integration of the trained CNN model into a comprehensive system capable of real-time video processing demonstrates the system’s applicability in diverse environments. Deployment considerations span various platforms, from centralized servers to edge devices, ensuring adaptability to different deployment scenarios. The existing systems are based on different methodologies but still, it is a challenging task as some of the factors like high speed of the vehicle, non-uniform vehicle number plate, the language of vehicle number and different lighting conditions can affect a lot in the overall recognition rate. Proposed system, driven by CNN technology, stands as a robust solution for accurate and efficient automated identification.

关键词：

来源：评论

学校读者我要写书评

暂无评论

Improving VANET Localization Performance with Data Verification: Challenges and Solutions

Improving VANET Localization Performance with Data Verificat...

引用

Decision Aid Sciences and Application (DASA), International Conference on

作者： Amjed Abbas Ahmed Aqeel Kamil Kadhim Ali Hamza Najim Khattab M. Ali Alheeti Nurhizam Safie Mohd Satar Aisa Hassan Abdallah Hashim Center for Cyber Security Faculty of Information Science and Technology Universiti Kebangsaan Malaysia Bangi Malaysia Department of Computer Techniques Engineering Imam AlKadhum College (IKC) Baghdad Iraq Department of Computer Technical Engineering Imam Al-Kadhum College (IKC) Baghdad Iraq Department of Computer Technical Engineering Imam Al-Kadhum College (IKC) Al-Diwaniyah Iraq Computer Networking Systems Department College of Computer and Information Technology University of Anbar Anbar Iraq Center for Cyber Security Faculty of Information Science and Technology Universiti Kebangsaan Malaysia (UKM) Bangi Malaysia Department of Electrical and Computer Engineering Faculty of Engineering International Islamic University Malaysia Selangor Malaysia

ISBN: (数字)9798350369106

ISBN: (纸本)9798350369113

In VANETs, the important and effective applications of vehicle localization include safety and communication applications. Thus, it is difficult to get very accurate localization in the dynamic and often very fluctuating environment of VANET. This research responds to this challenge with a keen thrust on improving vehicle localization using the Mobility Data Verification process. The primary goal of this work is to achieve high accuracy of vehicle positioning within VANET by checking the coherence of mobility data of moving vehicles. Hence, we present a concept that builds upon Mobility Data Verification to provide real-time location details from vehicles. Both GPS and onboard sensors feed this method, which uses a multi-level verification to flag out or eliminate the errors of deliberately spoofed data points. We use simulations to evaluate and validate the effectiveness of the proposed verification approach. As the proposed system, our previously developed and fine-tuned weighted localization (OWL) strategy does so. The performance evaluation results suggest that our proposed solution is feasible to distinguish automobiles that are not trustworthy and improve the performance of the localization algorithm.

关键词： Location awareness Resistance Performance evaluation Accuracy Vehicular ad hoc networks Real-time systems Sensors Safety Security Vehicle dynamics

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：