检索结果-内蒙古大学图书馆

Computing for Sustainability and Intelligent Future (COMP-SIF), International Conference on

作者： J L Mazher Iqbal Omprakash Gurrapu Saritha P S rajesh Kanna r r. Jayanthi Tejaskumar MB Department of Electronics and Communication Engineering Vel Tech Rangarajan Dr. Sagunthala R&D Institute of Science and Technology Avadi India Volvo Trucks North Carolina USA Department of Artificial intelligence and data science Dhanalakshmi Srinivasan College of Engineering Coimbatore India Department of Computer Science Christ University Bangalore India Department of Master of Computer Applications Dayananda Sagar College of Engineering Bengaluru India Don Bosco Institute of Technology Bangalore India

ISBN: (数字)9798331538538

ISBN: (纸本)9798331538545

The employ of deep learning methods for the diagnosis and prognosis model of chronic diseases is an important discovery to change the healthcare service. Some of the chronic diseases which prevalence and incidence rates remain high globally include diabetes, cardiovascular diseases, chronic kidney diseases, and cancers. There is nothing more critical than early diagnosis and accurate prediction of the patients' condition and the best course of action that has to be taken. This paper aims at examining the possibility of utilizing ANN, random Forest, XGBoost, and CNN to forecast the occurrence of the. due to integration of big and varied data which involve clinical characteristics, biochemical parameters and medical images among others, ML models have the ability recognize complex relations not easily recognizable by conventional diagnostic procedures. These illustrations prove that deep learning models or more specifically the convolutional neural networks for image diagnosis outperform other traditional methods in performance and prognosis. Nevertheless, some issues, such as data quality, model's interpretability, and its implementation into clinical practice, are still present. The challenges appeared in this paper are key to understanding the future of ML in healthcare as they can pave the way to the integration of such models into practice, therefore leading to early detection, better prognosis, and effective management of chronic diseases. This paper aims at exploring on how ML can be of significance in transformation of the health care sector and orderly improve patients care.

关键词： Neural networks Predictive models data models Chronic kidney disease diabetes Convolutional neural networks Cardiovascular diseases Prognostics and health management random forests Cancer

来源：评论

学校读者我要写书评

暂无评论

Potential of mm-Wave doppler-Polarimetric Profiler observations for Quality Assessment of Hydrometeor Classification Schemes

引用

Journal of Geophysical research: Atmospheres 2025年第11期130卷

作者： L. Bogerd H. Leijnse A. Overeem C. M. H. Unal r. Uijlenhoet S. van der Veen Hydrology and Environmental Hydraulics Group Wageningen University and Research Wageningen The Netherlands R&D Observations and Data Technology Royal Netherlands Meteorological Institute (KNMI) De Bilt The Netherlands Department of Water Management Faculty of Civil Engineering & Geosciences Delft University of Technology Delft The Netherlands Department of Geoscience and Remote Sensing Faculty of Civil Engineering & Geosciences Delft University of Technology Delft The Netherlands R&D Weather and Climate Modelling Royal Netherlands Meteorological Institute (KNMI) De Bilt The Netherlands

dual-polarization weather radars have improved the accuracy of precipitation estimates. However, challenges persist in evaluating hydrometeor classification (HMC) algorithms, thereby impacting the accuracy of precipitation estimates. This study proposes to use full doppler spectra in both polarizations from a Ka- and W-band doppler-polarimetric profiler with a 45° elevation angle to provide insights into hydrometeor characteristics. A novel methodology was developed to link the observed spectra with the output of an HMC scheme. We applied the wradlib HMC scheme using C-band weather radar data from the Netherlands for six cases (2021–2022). The HMC scheme output is used to calculate mixing ratios that are combined with the corresponding scattering properties using the Atmospheric radiative Transfer Simulator microwave single scattering properties database (frozen hydrometeors) and T-matrix calculations (liquid hydrometeors) to simulate doppler spectra of polarimetric variables that would be measured by the profiler. Comparing these simulations with actual profiler measurements enables a quality assessment. The method works in stratiform cases, but convective cases reveal the influence of turbulence and wind variability. Uncertainty arises from the selection of specific parameterizations for the particle size distribution and the relationship between hydrometeor size and terminal fall velocity as well as from the derived mixing ratios. Additionally, the 45° angle complicates separating horizontal wind from hydrometeor fall velocities, although the Mie notch in the dual-wavelength ratio can be effectively used to remove the radial wind component. Our results underline limitations that must be addressed but also show that integrating spectral and dual-frequency observations could yield valuable insights into hydrometeor characteristics. Combining spectral polarimetric observations and radar-based hydrometeor classification to gain insights into hydrometeor properties Co

关键词：

来源：评论

学校读者我要写书评

暂无评论

Learnable Convolutional Attention Network for Knowledge Graph Completion

SSRN

引用

SSrN 2023年

作者： Shang, Bin Zhao, Yinliang Liu, Jun School of Computer Science and Technology Xi’an Jiaotong University Shaanxi Xi’an710049 China Shaanxi Province Key Laboratory of Satellite and Terrestrial Network Tech. R&D Xi’an Jiaotong University Shaanxi Xi’an710049 China National Engineering Lab for Big Data Analytics Shaanxi Xi’an710049 China

recently, graph convolutional networks (GCNs) and graph attention networks (GATs) have been widely used in knowledge graph completion (KGC), which aims to solve the incompleteness of knowledge graphs (KGs). However, GCNs based KGC models tend to stack redundant information when one entity have several noisy neighbors. GATs based KGC models may focus on certain neighbors and lead to weakening of structural information. Furthermore, relations in KGs have their specific semantics and should be considered when aggregating neighbor information (message passing). To address the above limitations, we propose a learnable convolutional attention network for knowledge graph completion named LCA-KGC. LCA-KGC introduces a knowledge graph convolutional attention network by using the convolution operation before the attention mechanism to ensure the structural information acquisition and avoid the redundant information stacking. Furthermore, to complete the autonomous switching of GNNs types and eliminate the necessity of pre-validating the local structure of KGs, LCA-KGC designs a learnable knowledge graph convolutional attention network by comprising three types of GNNs in one learnable formulation. Moreover, a learnable message function is proposed to emphasize relational semantics when aggregating neighbor information. Extensive experiments on standard KG datasets validate the effectiveness of the proposed innovations, and LCA-KGC achieves state-of-the-art (SOTA) performance compared to existing approaches (e.g., compared to SOTA approaches, LCA-KGC improves Mrr from 0.360 to 0.372 on FB15k-237 dataset, and Hits@3 from 0.561 to 0.581 on YAGO3-10 dataset). © 2023, The Authors. All rights reserved.

关键词： Graph neural networks

来源：评论

学校读者我要写书评

暂无评论

Spectrum Aware Unequal Clustering with route Selection Protocol for Energy Efficient Cognitive radio Wireless Sensor Networks

Spectrum Aware Unequal Clustering with Route Selection Proto...

引用

Augmented Intelligence and Sustainable Systems (ICAISS), International Conference on

作者： Arthi. A Chinmaya Kumar Nayak Arunmurugan Subbaiyan M. Muthalakshmi Srimathi S Venkatesh p Department of Artificial Intelligence and Data Science Rajalakshmi Institute of Technology Chennai Faculty of Emerging Technologies Sri Sri University Sri Sri Vihar Odisha ECE Periyar Maniammai Institute of Science & Technology Thanjavur Tamilnadu India Vel Tech Rangarajan Dr. Sagunthala R&D Institute of Science and Technology Department of Bio Technology Saveetha School of Engineering SIMATS Chennai Tamilnadu India Department of ECE Ramco Institute of Technology Rajapalayam

Spectrum awareness with energy conservation is a main prerequisite in Cognitive radio Sensor Networks (CrSN) as the sensor nodes (SNs) operate in physically unattended areas and are energy constrained. Cluster based routing combined with channel allocation accomplishes energy effective path selection as a countermeasure to preserve energy. It is required to perform reliable routing with cluster based channel distribution contributing to development of network efficiency. Therefore, this study introduces a new Spectrum Aware metaheuristic Unequal Clustering with route Selection Protocol (SAMUC-rSP) technique for Energy Efficient CrSN. The focus of the SAMUC-rSP technique lies in the effective selection of CHs and routes in the CrSN. For accomplishing this, the SAMUC-rSP technique designs fruit fly optimization algorithm based unequal clustering (FFOA-UC) technique to produce unequal clusters in the CrSN. Besides, moth search optimization based routing (MSOr) approach is used to produce an optimal set of routes in the CrSN. The simulation results of the SAMUC-rSP technique are examined in terms of different measures in python environment. Extensive comparison analysis reported the superior results of the SAMUC-rSP method over other existing techniques in the CrSN.

关键词：

来源：评论

学校读者我要写书评

暂无评论

A dynamics Theory of Implicit regularization in deep Low-rank Matrix Factorization

SSRN

引用

SSrN 2023年

作者： Cao, Jian Qian, Chen Huang, Yihui Chen, dicheng Gao, Yuncheng dong, Jiyang Guo, di Qu, Xiaobo Department of Electronic Science Biomedical Intelligent Cloud R&D Center Fujian Provincial Key Laboratory of Plasma and Magnetic Resonance Xiamen University Xiamen361005 China School of Computer and Information Engineering Fujian Engineering Research Center for Medical Data Mining and Application Xiamen University of Technology Xiamen361024 China

Implicit regularization induced by gradient optimization is an important way to understand the generalization in neural networks. recent theory explains implicit regularization over the model of deep matrix factorization (dMF) and analyzes the trajectory of discrete gradient dynamics in the optimization process. These discrete gradient dynamics can mathematically characterize the practical learning rate of adaptive gradient optimization such as rMSProp. discrete gradient dynamics analysis has been successfully applied to shallow networks but encounters the difficulty of complex computation for deep networks. In this work, we introduce another discrete gradient dynamics approach to explain implicit regularization of rMSProp, i.e. landscape analysis. It mainly focuses on gradient regions like saddle points and local minima. We investigate that increasing learning rates benefit saddle point escaping (SPE) stages. In elucidating implicit regularization through the convergence of rMSProp, we prove that for a rank-r matrix reconstruction, dMF will converge to a second-order critical point after r stages of SPE. This conclusion is further experimentally verified on a low-rank matrix reconstruction problem. © 2023, The Authors. All rights reserved.

关键词： deep learning

来源：评论

学校读者我要写书评

暂无评论

Individual Privacy Protection in Cybersecurity Using data Analytics and Machine Learning

Individual Privacy Protection in Cybersecurity Using Data An...

引用

2024 International Conference on Innovative Computing, Intelligent Communication and Smart Electrical Systems, ICSES 2024

作者： Arun, M. Bhavadharani, K. Supreeth, B.r. Pundir, Sumit Pradeepa, K. razak, K. Abdul Vel Tech Rangarajan Dr. Sagunthala R&d Institute of Science and Technology Department of Cse Tamil Nadu Chennai India V.S.B. Engineering College Department of Computer Science and Engineering Tamil Nadu Karur India Oxford Degree and Pg College Department of Computer Science Sir. M. VisveshwaraiahLayout Ullal Bangalore India Graphic Era Deemed to be University Department of Computer Science and Engineering Uttarakhand Dehradun India Kpr College of Arts Science and Research Department of Artificial Intelligence and Data Science Tamil Nadu Coimbatore India K. Ramakrishnan College of Engineering Department of Mathematics Tamil Nadu Trichy India

ISBN: (纸本)9798331543617

Traditional cybersecurity addresses struggle to keep up with the ever-changing nature of cyber threats and usually fail to detect new privacy breaches. The paper proposes a proactive protection mechanism that employs data analytics and machine learning (ML) to promote individual privacy in the digital realm. The system analyzes several data sources, including system events, network traffic, and user activity logs, using advanced ML methods like random Forests (rF), Support Vector Machines (SVM), and Neural Networks (NN), to detect and prevent privacy violations. Ensemble learning techniques, feature engineering, and real-time monitoring enable adaptability in the face of shifting threats. According to the results, the proposed system surpassed the existing system in terms of accuracy (0.92), recall (0.89), precision (0.91), and F1 score (0.90). Its ability to detect unlawful access, unusual logins, and irregular data requests is proved by anomaly detection rates utilizing One-Class SVMs (91%) and Isolation Forest (89%). The system's low false positive rate of 5% demonstrates how effectively it can detect privacy breaches, resulting in stronger cybersecurity and greater privacy protection for both individuals and enterprises. © 2024 IEEE.

关键词： Cybersecurity

来源：评论

学校读者我要写书评

暂无评论

Evaluation and comparison of CMIP6 models and MErrA-2 reanalysis AOd against Satellite observations from 2000 to 2014 over China

引用

Geoscience Frontiers 2022年第2期13卷 165-185页

作者： Md.Arfan Ali Muhammad Bilal Yu Wang Zhongfeng Qiu Janet E.Nichol Gerrit de Leeuw Song Ke Alaa M.hawish Mansour Almazroui Usman Mazhar Birhanu Asmerom Habtemicheal M.Nazrul Islam Lab of Environmental Remote Sensing(LERS) School of Marine SciencesNanjing University of Information Science and TechnologyNanjing 210044China Department of Geography School of Global StudiesUniversity of SussexBrighton BN19RHUK Royal Netherlands Meteorological Institute(KNMI) R&D Satellite Observations3730AE De BiltThe Netherlands School of Atmospheric Physics Nanjing University of Information Science and TechnologyNanjing 210044China Aerospace Information Research Institute Chinese Academy of Sciences(AirCAS)No.20 Datun RoadChaoyang DistrictBeijing 100101China School of Environment Science and Spatial Informatics University of Mining and TechnologyXuzhouJiangsu 221116China Geological Survey of Jiangsu Province Nanjing 210018China Center of Excellence for Climate Change Research/Department of Meteorology King Abdulaziz UniversityJeddah 21589Saudi Arabia Climatic Research Unit School of Environmental SciencesUniversity of East AngliaNorwichUK School of Remote Sensing and Geomatics Engineering Nanjing University of Information Science and TechnologyNanjing 210044China Department of Physics Wollo UniversityDessie P.O.Box 1145Ethiopia

rapid industrialization and urbanization along with a growing population are contributing significantly to air pollution in *** of long-term aerosol optical depth(AOd)data from models and reanalysis,can greatly promote understanding of spatiotemporal variations in air pollution in *** do this,AOd(550 nm)values from 2000 to 2014 were obtained from the Coupled Model Intercomparison Project(CIMP6),the second version of Modern-Era retrospective analysis for research,and Applications(MErrA-2),and the Moderate resolution Imaging Spectroradiometer(MOdIS;flying on the Terra satellite)combined dark Target and deep Blue(dTB)aerosol *** used the TerraMOdIS dTB AOd(hereafter MOdIS dTB AOd)as a standard to evaluate CMIP6 Ensemble AOd(hereafter CMIP6 AOd)and MErrA-2 reanalysis AOd(hereafter MErrA-2 AOd).results show better correlations and smaller errors between MErrA-2 and MOdIS dTB AOd,than between CMIP6 and MOdIS dTB AOd,in most regions of China,at both annual and seasonal ***,significant under-and over-estimations in the MErrA-2 and CMIP6 AOd were also observed relative to MOdIS dTB *** long-term(2000-2014)MOdIS dTB AOd distributions show the highest AOd over the North China Plain(0.71)followed by Central China(0.69),Yangtse river delta(0.67),Sichuan Basin(0.64),and Pearl river delta(0.54)*** lowest AOd values were recorded over the Tibetan Plateau(0.13±0.01)followed by Qinghai(0.19±0.03)and the Gobi desert(0.21±0.03).Large amounts of sand and dust particles emitted from natural sources(the Taklamakan and Gobi deserts)may result in higher AOd in spring compared to summer,autumn,and *** were also calculated for 2000-2005,for2006-2010(when China introduced strict air pollution control policies during the 11 th Five Year Plan or FYP),and for 2011-2014(during the 12 th FYP).An increasing trend in MOdIS dTB AOd was observed throughout the country during *** uncontrolled industrialization,urbanization,and rapid eco

关键词： Aerosol Optical depth Air pollution MErrA-2 MOdIS Air quality

来源：评论

学校读者我要写书评

暂无评论

A Novel Architecture for Efficient Mobile Communication based on Scattered and diverse Combination using Prioritize and Postoperative Soft Modulation

A Novel Architecture for Efficient Mobile Communication base...

引用

International Conference on ent Computing and Control Systems (ICICCS)

作者： J. Sumithra M Mercy Theresa S. Geetha Priya Beulah Jackson V. Samuthira Pandi A. Balamurugan Department of Electronics and Communication Engineering R. M. D Engineering College Tamil Nadu India Department of Data Science and Business Systems SRM Institute of Science and Technology Tamil Nadu India Department of Computer Science and Engineering R. M. D Engineering College Tamil Nadu India Department of Electronics and Communication Engineering Vel Tech Rangarajam Dr. Sagunthala R&D Institute of Science and Technology Tamil Nadu India Department of Electronics and Communication Engineering SRM Institute of Science and Technology Tamil Nadu India Department of Electrical and Electronics Engineering Chennai Institute of Technology Tamil Nadu India

The proposed work suggests a novel approach for the cooperation of geographically scattered individuals and the blending of different points of view. This work ensures that the center on heterogeneous networks in which nodes support both long-range low-rate radio Frequency (rF) interfaces (like cellular) and short-range high-rate rF interfaces (such as IEEE802.11) that communicate over urban rayleigh fading channels. These interfaces are used to connect the devices. The interfaces are what are used to link the various gadgets. This research work proposes and evaluates a wide variety of hierarchically distributed cooperation strategies that can function despite resource restrictions such as limited short-range rF bandwidth. These strategies are contained within the confines of this paradigm, which serves as the framework for proposed work. Post Soft-demodulation Combining, often referred to as PSdC, and Priority Maximum ratio Combining are the two approaches that come highly recommended by our team (PMrC). This work provides evidence that using the offered methods increases throughput, reliability, and signal-to-noise ratio. Moreover, this study found that PMrC and PSdC have the ability to improve throughput performance by more than an order of magnitude across a wide range of communication circumstances.

关键词：

来源：评论

学校读者我要写书评

暂无评论

Photovoltaic Powered Electric Vehicle Using direct Torque Control and IoT Monitoring

Photovoltaic Powered Electric Vehicle Using Direct Torque Co...

引用

2023 International Conference on Power Energy, Environment and Intelligent Control, PEEIC 2023

作者： Sathyanathan, P. Manikandan, S. Chandrakala, d. raian, V.r. Ishwarya, M.V. Sujatha, S. VelTech Rangarajan Dr. Sagunthala R&d Institute of Science and Technology Department of Electrical and Electronics Engineering Tamil Nadu Chennai India Meenakshi Sundararajan Engineering College Department of Electrical and Electronics Engineering Tamil Nadu Chennai India Easwari Engineering College Department of Electrical and Electronics Engineering Tamil Nadu Chennai India Mvj College of Engineering Department of Electrical and Electronics Engineering Karnataka Bangalore India Agni College of Technology Artificial Intelligence and Data Science Department Tamil Nadu Thazhambur India Saveetha School of Engineering Saveetha Institute of Medical and Technical Sciences Saveetha University Department of Biomedical Engineering Tamil Nadu Chennai India

ISBN: (纸本)9798350357769

The concept of speed control of a Permanent Magnet Synchronous Motor (PMSM) using direct Torque Control (dTC) is proposed for use in electric vehicles (EVs). A solar Photo Voltaic (PV) system powers the PMSM, and they are monitored using Internet of Things (IoT). In this configuration, a PMSM serves as the EV's main propulsion engine, with speed control achieved by dTC which provides fast and accurate torque and flux control. By using dTC to accurately control the PMSM speed, EV operation may be more reliable and efficient. This integration lessens the dependence on conventional energy sources and enables the functioning of the EV in a sustainable and ecologically responsible manner. The PV system offers a sufficient and clean source of energy, enhancing the EV's overall performance and range. IoT technology is used to keep monitor the motor speed and data from the PV system. The IoT allows instantaneous information exchange and connection across systems and objects. © 2023 IEEE.

关键词： direct torque control Electric vehicle Internet of things and Permanent magnet synchronous motor Photovoltaic

来源：评论

学校读者我要写书评

暂无评论

Hard Sample Matters a Lot in Zero-Shot Quantization

Hard Sample Matters a Lot in Zero-Shot Quantization

引用

Conference on Computer Vision and Pattern recognition (CVPr)

作者： Huantong Li Xiangmiao Wu Fanbing Lv daihai Liao Thomas H. Li Yonggang Zhang Bo Han Mingkui Tan South China University of Technology Information Technology R&D Innovation Center of Peking University Changsha Hisense Intelligent System Research Institute Co. Ltd School of Electronic and Computer Engineering Peking University Shenzhen Graduate School Shenzhen China Hong Kong Baptist University Key Laboratory of Big Data and Intelligent Robot Ministry of Education PengCheng Laboratory

Zero-shot quantization (ZSQ) is promising for compressing and accelerating deep neural networks when the data for training full-precision models are inaccessible. In ZSQ, network quantization is performed using synthetic samples, thus, the performance of quantized models depends heavily on the quality of synthetic samples. Nonetheless, we find that the synthetic samples constructed in existing ZSQ methods can be easily fitted by models. Accordingly, quantized models obtained by these methods suffer from significant performance degradation on hard samples. To address this issue, we propose HArd sample Synthesizing and Training (HAST). Specifically, HAST pays more attention to hard samples when synthesizing samples and makes synthetic samples hard to fit when training quantized models. HAST aligns features extracted by full-precision and quantized models to ensure the similarity between features extracted by these two models. Extensive experiments show that HAST significantly outperforms existing ZSQ methods, achieving performance comparable to models that are quantized with real data.

关键词：

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：