检索结果-内蒙古大学图书馆

3rd International Conference on Machine Learning and Data engineering, ICMLDE 2024

作者： Puneetha, B.H. Kumar, Manoj M.V. Prashanth, B.S. Kumara, Ajay M.A. Department of Computer Science and Business Systems Bapuji Institute of Engineering and Technology Karnataka Davanagere India Department of Information Science and Engg Nitte Meenakshi Institute of Technology Karnataka Bangalore560064 India Department of Computer Science University of North Carolina Wilmington United States

Recent advancements in handling class imbalance include hybrid approaches that combine data-level and algorithmic-level techniques. However, challenges such as computational efficiency and adaptability to dynamic datasets remain to be prominently addredssed. This research paper introduces novel hybrid techniques that combine both data-level and algorithmic-level approaches to address class imbalance in large datasets. These techniques are designed to improve class distribution and processing efficiency while enhancing prediction accuracy, particularly for minority classes. The proposed model applies K-means SMOTE for over-sampling the minority class and centroid-based undersampling for the majority class, followed by Classification using XGBoost. Evaluations on multiple datasets demonstrate significant improvements in precision, recall, AUC-ROC, and F1-score compared to traditional methods. In addition to enhancing robustness and fairness, the approach effectively addresses challenges related to scalability, data diversity, and dynamic data sources in big data analytics. The hybrid technique achieved a 10-20% improvement in key performance metrics, including precision, recall, F1-score, and AUC-ROC, across various UCI datasets such as Diabetes, Hepatitis, and German Credit, confirming the model's superiority in handling class imbalance and improving overall Classification performance. © 2025 The Author(s).

关键词： Large datasets

来源：评论

学校读者我要写书评

暂无评论

Fault-tolerant Mobile Service Offloading in Mobile Edge Computing

引用

IEEE Transactions on Automation science and engineering 2025年

作者： Long, Tingyan Xia, Yunni Zhou, Mengchu Li, Jianqi Ma, Yong Ai-Turki, Yusuf Guizhou University School of Computer Science and Technology Guiyang550025 China Chongqing University College of Computer Science Chongqing400030 China New Jersey Institute of Technology Department of Electrical and Computer Engineering NewarkNJ07102 United States Global energy interconnection research institute Co. Ltd Beijing102209 China Jiangxi Normal University School of Computer and Information Engineering Nanchang China King Abdulaziz University K. A. Care Energy Research and Innovation Center The Center of Research Excellence in Renewable Energy and Power Systems Department of Electrical and Computer Engineering Faculty of Engineering Jeddah21589 Saudi Arabia

Mobile edge computing (MEC) is an evolving paradigm for rendering services through network-accessible resources deployed over Internet of Things (IoT) nodes at the edge. Nevertheless, an MEC environment usually employs thousands of physical machines connected via hundreds of switches/routers that communicate and coordinate to deliver computing service. In such complicated systems, faults caused by software, human errors, and hardware are often unavoidable. The edge of network presents a dynamic environment with great quantities of terminals, high mobility of mobile devices, heterogeneous applications, and intermittent traffic. In such an environment, MEC can suffer from unbalanced resource provisioning and interruptions of faults occurring at different levels, which further causes task faults and affects service quality. To address this challenge, this work proposes a novel fault-tolerant offloading method for handling faults by leveraging a reinforcement-learning-based service offloading decision model. The model synthesizes a Dueling Deep Q Network (DQN)-based algorithm for deciding user offloading behaviors and an adaptive checkpointing method for improving task execution reliability. For the purpose of model validation and comparison, extensive simulations are conducted. Numerical results clearly demonstrate that the proposed method is highly effective and outperforms existing methods. © 2004-2012 IEEE.

关键词： Mobile applications

来源：评论

学校读者我要写书评

暂无评论

Performance Analysis of Tractors Traveling on Rough Terrain

引用

ASME Letters in Dynamic systems and Control 2022年第4期2卷 041001页

作者： Izumi, Shinsaku Hayashida, Riku Xin, Xin Faculty of Computer Science and Systems Engineering Okayama Prefectural University Okayama 719-1197 Japan Graduate School of Computer Science and Systems Engineering Okayama Prefectural University Okayama 719-1197 Japan

This article considers tractors traveling on rough terrain and studies the effects of the road surfaces on their behavior. Clarifying these effects is useful for their stable traveling on rough terrain, which contributes to the automation of tractors. We select the H2norm as a performance index to quantify the behavior of tractors and attempt to analyze the norm of a system corresponding to a tractor traveling on rough terrain. To improve the accuracy of the analysis, we modify the system by focusing on the relationship between the H2norm, white noise, and the characteristics of the road surface. Subsequently, the H2norm of the resulting system is derived as a function of the parameters, including the road roughness. Numerical examples demonstrate that the analysis result based on the modified system accurately captures the behavior of the tractor compared with the result based on the original system.

关键词： linear systems vibrations vehicle dynamics and control Design Pavement Surface roughness Roads Vibration White noise

来源：评论

学校读者我要写书评

暂无评论

A Storage Optimization Scheme for Blockchain Transaction Databases

引用

computer systems science & engineering 2021年第3期36卷 521-535页

作者： Jingyu Zhang Siqi Zhong Jin Wang Xiaofeng Yu Osama Alfarraj School of Computer and Communication Engineering Changsha University of Science and TechnologyChangsha410114China School of Systems Engineering National University of Defense TechnologyChangsha410000China School of Information Science and Engineering Fujian University of TechnologyFuzhou350000China School of Business Nanjing UniversityNanjing210093China Computer Science Department Community CollegeKing Saud UniversityRiyadh11437Saudi Arabia

As the typical peer-to-peer distributed networks, blockchain systemsrequire each node to copy a complete transaction database, so as to ensure newtransactions can by verified independently. In a blockchain system (e.g., bitcoinsystem), the node does not rely on any central organization, and every node keepsan entire copy of the transaction database. However, this feature determines thatthe size of blockchain transaction database is growing rapidly. Therefore, with thecontinuous system operations, the node memory also needs to be expanded tosupport the system running. Especially in the big data era, the increasing networktraffic will lead to faster transaction growth rate. This paper analyzes blockchaintransaction databases and proposes a storage optimization scheme. The proposedscheme divides blockchain transaction database into cold zone and hot zone usingexpiration recognition method based on Least Recently Used (LRU) algorithm. Itcan achieve storage optimization by moving unspent transaction outputs outsidethe in-memory transaction databases. We present the theoretical analysis on theoptimization method to validate the effectiveness. Extensive experiments showour proposed method outperforms the current mechanism for the blockchaintransaction databases.

关键词： Blockchain distributed systems transaction databases

来源：评论

学校读者我要写书评

暂无评论

The performance of circularly polarized phased sub-array antennas for 5G laptop devices investigating the radiation effects

引用

Progress in Electromagnetics Research C 2021年 110卷 267-283页

作者： Mahmoud, Korany R. Baz, Abdullah Alhakami, Wajdi Alhakami, Hosam Montaser, Ahmed M. Department of Electronics and Communications Faculty of Engineering Helwan University Cairo Egypt National Telecommunications Regulatory Authority Ministry of Communication and Information Technology Giza Egypt Computer Engineering Department College of Computer and Information Systems Umm Al-Qura University Makkah Saudi Arabia Department of Computer Science College of Computer and Information Technology Taif University P. O. Box 11099 Taif21944 Saudi Arabia Department of Computer Science College of Computer and Information Techonogy Umm Al-Qura University Makkah21955 Saudi Arabia Electrical Engineering Department Sohag University Sohag Egypt

In this paper, the performance of circularly polarized (CP) adaptive sub-arrays integrated into 5G laptop device is investigated in the presence of a whole-body human phantom model. In addition, the radiation effect of the steered beam patterns has been analyzed by calculating the specific absorption rate distribution and temperature rise. In this target, a single-feed CP antenna element has been firstly designed to resonate at 28GHz with high realized gain and radiation efficiency. Then, 4 subarrays have been constructed in a rectangular configuration with four-elements for each sub-array. To let the study more realistic, a complete human model is considered to investigate the radiation effects. The measured reflection coefficient and realized gain results of the designed antenna element are found to be -30 dB and 7.82 dB, respectively, in the assigned frequency band. Likewise, the antennas sub-arrays have approximately kept the same impedance matching attitude with high insertion loss of -22 dB and a realized gain and radiation efficiency of 16.85 dB and 86%, respectively, on average. Furthermore, the sub-arrays scan patterns and coverage efficiency have been studied considering the existence of the human body in different scenarios. Regarding the RF exposure, the results show that the resultant maximum values of specific absorption rate and power density do not exceed 1.52 W/kg and 3.5W/m2, respectively, whereas the maximum exposure temperature in such a case is less than 2.8°C after 30 minutes and decreases to 0.5°C after a penetration depth of 3mm which reflects the possibility of safe use. © 2021, Electromagnetics Academy. All rights reserved.

关键词： Radiation efficiency

来源：评论

学校读者我要写书评

暂无评论

Enhancing Sentiment Analysis on SEED-IV Dataset with Vision Transformers: A Comparative Study 23

Enhancing Sentiment Analysis on SEED-IV Dataset with Vision ...

引用

Proceedings of the 2023 11th International Conference on Information Technology: IoT and Smart City

作者： Imad Eddine Tibermacine Ahmed Tibermacine Walid Guettala Christian NAPOLI Samuele Russo Department of Computer Automation and Management Engineering Sapienza Unviersity of Rome Italy Department of Computer Science University of Biskra Algeria Department of Computer Automation and Management Engineering Sapienza Unviersity of Rome Italy and Institute for Systems Analysis and computer Science Italian National Research Council Italy and Department of Computational Intelligence Czestochowa University of Technology Poland Department of Psychology Sapienza Unviersity of Rome Italy

ISBN: (纸本)9798400709043

This paper introduces a new approach to emotion classification utilising deep learning models, specifically the Vision Transformer (ViT) model, in the analysis of electroencephalogram (EEG) signals. A dual-feature extraction approach was implemented in our study, utilising Power Spectral Density and Differential Entropy, to analyse the SEED IV dataset. This methodology resulted in the detailed classification of four distinct emotional states. The ViT model, which was originally designed for image processing, has been successfully applied to EEG signal analysis. It demonstrated remarkable performance by attaining a test accuracy of 99.02% with little variance. Notably, it outperformed conventional models like GRUs, LSTMs, and CNNs in this context. The findings of our study indicate that the ViT model has a high level of effectiveness in accurately identifying complex patterns present in EEG data. Specifically, the precision and recall rates achieved by the model surpass 98%, while the F1 score is estimated to be about 98.9%. The results of this study not only demonstrate the efficacy of transformer-based models in analysing cognitive states, but also indicate their considerable potential in improving systems for sympathetic human-computer interaction.

关键词： Classification Deep Learning EEG Signal Analysis Vision Transformer

来源：评论

学校读者我要写书评

暂无评论

A Novel Approach to Menstrual Cycle Prediction Using GAN-Generated Data and Transformer Model for Tabular Data

A Novel Approach to Menstrual Cycle Prediction Using GAN-Gen...

引用

2024 IEEE International Women in engineering (WIE) Conference on Electrical and computer engineering, WIECON-ECE 2024

作者： Ahamed, Md Younus Syed, Md Asif Bin Tani, Mafuza Nasrin Tasnim, Tanpia Habib, Al Zadid Sultan Bin West Virginia University Lane Department of Computer Science and Electrical Engineering MorgantownWV26506 United States West Virginia University Department of Industrial and Management Systems Engineering MorgantownWV26506 United States Army Medical College Rangpur Rangpur5400 Bangladesh Green University of Bangladesh Narayanganj Dhaka1461 Bangladesh

ISBN: (纸本)9798331535476

Accurate menstrual cycle prediction is crucial for women's health and fertility management. While prior studies have utilized machine learning models to predict cycle length and classify regularity, they often struggled with the low Observationto-Variable Ratio (OVR) inherent in menstrual cycle data. This study tackles this issue by employing Generative Adversarial Networks (GANs) to generate synthetic data and Transformerbased models for both classification a nd r egression t asks. We evaluated multiple machine learning models for menstrual cycle phase classification a nd l ength p rediction. F or t he classification of regular versus irregular cycles, TabNet significantly outperformed baseline models, including Logistic Regression, Random Forest, and LSTM, achieving 99.67% accuracy, an AUC of 0.96, an F1 score of 0.99, 92.37% specificity, a nd 99.66% sensitivity, with CTGAN-generated synthetic data enhancing its performance. In the regression task of predicting cycle length, TabNet demonstrated strong predictive power, with an MSE of 1.29, RMSE of 1.13, and an R2 score of 0.87. Although Random Forest performed slightly better with an MSE of 0.958 and an R2 score of 0.90, TabNet remained competitive. These results highlight the potential of GAN-generated data and advanced models like TabNet in overcoming low OVR challenges and improving menstrual cycle prediction. Future research may incorporate personalized health data to further enhance model accuracy and generalizability. © 2024 IEEE.

关键词： Prediction models

来源：评论

学校读者我要写书评

暂无评论

Manipulating the Posterior Support of a Discrete Bayesian Estimator Under Full Sensor Control

引用

IFAC-PapersOnLine 2023年第2期56卷 252-257页

作者： Daniel Selvaratnam Farhad Farokhi Iman Shames Henrik Sandberg Decision and Control Systems School of Electrical Engineering & Computer Science KTH Royal Institute of Technology 114 28 Stockholm Sweden Department of Electrical and Electronic Engineering The University of Melbourne Parkville VIC 3010 Australia College of Engineering & Computer Science Australian National University Canberra ACT 2601 Australia

The asymptotic implausibility problem is introduced from the perspective of an adversary that seeks to drive the belief of a recursive Bayesian estimator away from a particular set of parameter values. It is assumed that the adversary controls all sensors informing the estimator, and can transmit false measurements stochastically according to a fixed distribution of its choice. First, we outline a method for verifying whether a given distribution solves the problem. We then consider the class of spoofing attacks, and show that the asymptotic implausibility problem has a solution if and only if it can be solved by a spoofing attack. Attention is restricted to finite parameter and observation spaces.

关键词： Network Security Bayesian Estimation Cyber-Physical systems Network systems Resilient Control systems Signal Processing

来源：评论

学校读者我要写书评

暂无评论

An Architecture-based Self-Typing Service for Cloud Native Applications

An Architecture-based Self-Typing Service for Cloud Native A...

引用

Smart Electronics and Communication (ICOSEC), International Conference on

作者： Srinivasa Rao Mandalapu Kalaimagal Sivamuni Chitra Devi D Upendra Singh Aswal S. Irin Sherly N. Alangudi Balaji Department of Computer Science and Business Systems R.V.R. & J.C. College of Engineering Chowdavaram India Department of Computer Science and Engineering MK College of Engineering and Technology Puduvoyal India Department of Computer Science and Engineering S.A.Engineering college Thiruverkadu India Department of Computer Science & Engineering Graphic Era Deemed to be University Dehradun India Department of Information Technology Panimalar Engineering College Poonamallee India Department of Computer Science and Engineering Koneru Lakshmaiah Education Foundation Vaddeswaram India

This research study analyzes Kubow, a service that automatically adjusts to new circumstances, built on a modular framework for cloud-native applications. To accommodate Kubernetes and Docker containers, the Rainbow self-adaptation framework was upgraded to become Kubow. This research provides an overview of Kubow's architecture, highlights its main design decisions, and illustrates it with a simple example of how to set it up and utilize it. It takes more than just installing software on virtual machines to get apps running in the cloud. Cloud applications need continuous management so that they can 1) adjust their resources according to the demand, and 2) respond to temporary challenges by duplicating and restarting the components to offer resilience on unstable infrastructure. To give programmed and responsive responses to disappointments (wellbeing the board) and changing natural conditions (auto-scaling), and framework boundaries.

关键词：

来源：评论

学校读者我要写书评

暂无评论

Robust Online Multiband Drift Estimation in Electrophysiology Data

Robust Online Multiband Drift Estimation in Electrophysiolog...

引用

International Conference on Acoustics, Speech, and Signal Processing (ICASSP)

作者： Charlie Windolf Angelique C. Paulk Yoav Kfir Eric Trautmann Domokos Meszéna William Muñoz Irene Caprara Mohsen Jamali Julien Boussard Ziv M. Williams Sydney S. Cash Liam Paninski Erdem Varol Departments of Statistics Zuckerman Institute Columbia University Departments of Neurology Center for Neurotechnology and Neurorecovery Massachusetts General Hospital Harvard Medical School Departments of Neurosurgery Departments of Computer Science & Engineering New York University

High-density electrophysiology probes have opened new possibilities for systems neuroscience in human and non-human animals, but probe motion poses a challenge for downstream analyses, particularly in human recordings. We improve on the state of the art for tracking this motion with four major contributions. First, we extend previous decentralized methods to use multiband information, leveraging the local field potential (LFP) in addition to spikes. Second, we show that the LFP-based approach enables registration at sub-second temporal resolution. Third, we introduce an efficient online motion tracking algorithm, enabling the method to scale up to longer and higher-resolution recordings, and possibly facilitating real-time applications. Finally, we improve the robustness of the approach by introducing a structure-aware objective and simple methods for adaptive parameter selection. Together, these advances enable fully automated scalable registration of challenging datasets from human and mouse.

关键词： Tracking Signal processing algorithms Estimation Electrophysiology Robustness Real-time systems Mice

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：