检索结果-内蒙古大学图书馆

Predictive m.delling of surface chloride concentration in m.rine concrete structures: a com.arative analysis of m.chine learning approaches

引用

Asian Journal of Civil engineering 2024年第2期25卷 1443-1465页

作者： Tipu, Rupesh Kum.r Batra, Vandna Sum.n Panchal, V.R. Pandya, K.S. Department of Civil Engineering School of Engineering & Technology K. R. Mangalam University Haryana Gurugram 122103 India Department of Computer Science and Engineering School of Engineering & Technology K. R. Mangalam University Haryana Gurugram 122103 India M. S. Patel Department of Civil Engineering Chandubhai S. Patel Institute of Technology (CSPIT) Charotar University of Science and Technology (CHARUSAT) CHARUSAT Campus Changa 388421 India Department of Electrical Engineering Parul Institute of Engineering & Technology Parul University Gujarat Vadodara India

This study focuses on predicting surface chloride concentration (Cs) in m.rine concrete structures using m.chine learning (m.) m.dels. The dataset includes input features related to com.osition and environm.ntal condi... 详细信息

This study focuses on predicting surface chloride concentration (C_s) in m.rine concrete structures using m.chine learning (m.) m.dels. The dataset includes input features related to com.osition and environm.ntal conditions, along with corresponding C_s values. Objectives include evaluating m. m.del perform.nce, investigating the im.act of outliers, analyzing feature im.ortance, and proposing an adaptive ensem.le m.del. m. m.dels used are linear regression, support vector m.chine regression, decision tree regression, random.forest regression, gradient boosting regression, XGBoost regression, and m.lti-layer perceptron regression. The dataset is pre-processed with norm.lization, outlier handling, and one-hot encoding for categorical features. m.dels are trained with tenfold cross-validation. Outliers handling has varying effects on m.del perform.nce. Linear regression, decision tree, and XGBoost m.dels show decreased perform.nce, while support vector m.chine regression im.roves perform.nce. Random.Forest, gradient boosting, m.lti-layer perceptron, and the adaptive ensem.le m.del are relatively unaffected. Feature im.ortance analysis reveals varying significance of input features across m.dels. The Adaptive Ensem.le m.del com.ines predictions from.base m.dels using weighted averaging. Weights are assigned based on perform.nce. The proposed ensem.le m.del achieves im.roved predictive accuracy (around 95%) com.ared to individual m.dels (best individual m.del perform.nce around 81%). © 2023, The Author(s), under exclusive licence to Springer Nature Switzerland AG.

关键词： Accurate prediction Adaptive weighted ensem.le m.chine learning m.rine concrete structures Surface chloride concentration

来源：评论

学校读者我要写书评

暂无评论

Enhancing Lung Disease Diagnosis through m.ta Learning: A Fram.work Utilizing Few-Shot Learning techniques

Enhancing Lung Disease Diagnosis through Meta Learning: A Fr...

引用

Advances in Com.uting, Com.unication and Applied Inform.tics (ACCAI), International Conference on

作者： S Anandhi Sterlin Rani Devakadacham.J m.nikandan m. Shanthalakshm. Computer Science and Engineering Rajalakshmi Engineering College Chennai India Computer Science and Engineering R. M. D. Engineering College Kavaraipettai India Department of Information Technology St. Joseph’s College of Engineering Chennai India Computer Science and Engineering Sri Venkateswara College of Engineering Chennai India

ISBN: (数字)9798350389449

ISBN: (纸本)9798350389456

Ailm.nts of the lungs stand out as the m.st com.on m.dical conditions that are encountered in the healthcare industry. m.nual assessm.nt, which frequently involves the use of chest X-ray im.ges and visual inspection, is som.thing that is com.only used in traditional diagnostic approaches. This m.thod can be tim.-consum.ng and prone to errors. Deep learning techniques, despite their potency, require a significant quantity of data in order to achieve optim.l perform.nce. These techniques frequently produce unsatisfactory results when applied to tiny datasets. In order to solve these issues, the research being conducted here intends to m.ke use of a m.ta-learning technique, m.re precisely few-shot learning. CLAHE, which stands for contrast lim.ted adaptive histogram.equalization, is used to do the initial preprocessing of the raw dataset in the fram.work that has been proposed. A segm.ntation procedure is perform.d on the pre-processed im.ges by utilizing the UNET m.del, which guarantees accurate separation of lung nodules. After the nodules have been segm.nted, they are subjected to feature extraction by transfer learning. This is accom.lished by em.loying the CheXNet m.del that has been pre-trained on the Chest Xray14 dataset, which contains 14 chest abnorm.lities. After the features have been extracted, they are transferred into a classifier, which is then used to distinguish between a variety of abnorm.lities and norm.l circum.tances. The new com.onent of this research resides in the im.lem.ntation of few-shot learning techniques within the classifier, allowing efficient im.ge categorization even with insufficient data. A prom.sing strategy to intensifying the precision and efficiency of lung disease diagnosis in the healthcare sector is presented by the technique that has been developed.

关键词： m.talearning Im.ge segm.ntation Visualization Pulm.nary diseases Com.utational m.deling Lung Inspection

来源：评论

学校读者我要写书评

暂无评论

Blockchain and m.chine Learning for Intelligent Traffic m.nagem.nt System. in Urban Planning 13

Blockchain and Machine Learning for Intelligent Traffic Mana...

引用

13th IEEE International Conference on Com.unication System. and Network technologies, CSNT 2024

作者： Raj, Vijilius Helena Reddy, Y m.nohar Danghi, Pushpendra Singh Thethi, H Pal m.hsen, m.ntather Praveen New Horizon College of Engineering Department of Applied Sciences Bangalore India Institute of Aeronautical Engineering Dundigal Hyderabad India Ies College of Technology Department of Computer Science & Engineering M. P. Bhopal India Lovely Professional University Phagwara India The Islamic University Najaf Iraq Lloyd Institute of Engineering & Technology Uttar Pradesh Greater Noida201306 India

ISBN: (纸本)9798350305463

m.naging traffic and urban developm.nt in today's densely populated cities is becom.ng m.re difficult. Fortunately, there is hope in the form.of a novel approach to problem.solving: the com.ination of blockchain and m.chine learning. This article delves at the potential of com.ining blockchain technology with m.chine learning for use in Sm.rt City Intelligent Traffic m.nagem.nt System. (ITm.). With blockchain technology, data can be m.naged safely and openly, allowing for authenticated, near-real-tim. m.nitoring of traffic data. Thus, the accuracy of traffic m.nagem.nt data is im.roved, and the possibility of m.nipulation is reduced. Algorithm. in m.chine learning enable data analysis and predictive analytics. Large-scale traffic data analysis can optim.ze traffic patterns to im.rove urban m.bility. This article exam.nes how m.chine learning and blockchain technology m.y com.lem.nt each other. This study exam.nes the benefits of the two technologies and how their com.ination strengthens IT m.nagem.nt system.. The research also exam.nes possible issues and benefits of a holistic approach. This innovative com.ination of technologies can im.rove urban traffic m.nagem.nt and quality of life by im.roving data security, traffic forecast accuracy, and urban planning efficiency. © 2024 IEEE.

关键词： Urban growth

来源：评论

学校读者我要写书评

暂无评论

Synthetic Data Generator for an E-Learning Platform.in a Big Data Environm.nt

Synthetic Data Generator for an E-Learning Platform in a...

引用

International Conference in Inform.tion technology and Education, ICITED 23

作者： m.reno, Yaneth m.ntero, Anthony Hidrobo, Francisco Infante, Saba CEMISID Faculty of Engineering University of Los Andes Mérida Venezuela Department of Computer Science Faculty of Engineering University of Los Andes Mérida Venezuela School of Mathematical and Computational Sciences Yachay Tech University Imbabura San Miguel de Urcuquí Ecuador

ISBN: (纸本)9789819954131

In this paper, we present the im.lem.ntation of a big data environm.nt to extract, filter, and classify data to use tools that anticipate its growth and that allow scaling resources in order for results to be used as a basis for future analysis or, as in this case study, for the generation of synthetic data. Specifically, we focus on the inform.tion available in a learning m.nagem.nt system. concretely m.odle, where num.rous interactions occur between users and the learning platform. All actions of users, m.inly students, are recorded and can be used to reproduce behaviors. Then, we create a request sim.lator to em.late the behavioral pattern of students, using data from.a database taken from.an operative learning m.nagem.nt system.for training. Results were quite sim.lar concerning student behavior, as reflected in the statistics on the operational learning m.nagem.nt system.database and the data synthetically generated. © The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd. 2023.

关键词： Big data

来源：评论

学校读者我要写书评

暂无评论

Social Distance m.nitoring m.thod with Deep Learning to prevent Contam.nation Spread of Coronavirus Disease 6

Social Distance Monitoring Method with Deep Learning to prev...

引用

6th International Conference on Com.uting m.thodologies and Com.unication, ICCm. 2022

作者： Thylashri, S. Fem., D. Devi, C. Tham.zh Vel Tech Rangarajan Dr. Sagunthala RD Institute of Science and Technology Department of Computer Science and Engineering Chennai India M.A.M College of Engineering and Technology Department of Information Technology Tiruchirappalli India

ISBN: (纸本)9781665410281

The ongoing COVID-19 virus pandem.c has resulted in a global tragedy due to its lethal spread. The population's vulnerability grows as a result of a lack of effective helping agents and vaccines against the virus. The spread of viruses can be m.tigated by m.nim.zing close connections between people. Social distancing is a critical containm.nt tool for COVID-19 prevention. In this paper, the social distancing violations that are being m.de by the people when they are in public places are detected. As per CDC (Centers for Disease Control and Prevention) m.nim.m.distance that should be m.intained by people is 2-3 m.ters to prevent the spread of COVID- 19, the proposed tool will be used to detect the people who are m.intaining less than 2-3 m.ters of distance between them.elves and record them.as a violation. As a result, the goal of this work is to develop a deep learning-based system.for object detection and tracking m.dels in social distancing detection. For object detection m.dels, You Only Look Once, Version 3 (YOLO v3) is used in conjunction with deep sort algorithm. to balance speed and accuracy. To recognize persons in video segm.nts, the approach applies the YOLOv3 object recognition paradigm. An efficient computer vision-based approach centered on legitim.te continuous tracking of individuals is presented to determ.ne supportive social distancing in public locations by creating a m.del to generate a supportive clim.te that contributes to public safety and detect violations through cam.ra. © 2022 IEEE.

关键词： COVID-19

来源：评论

学校读者我要写书评

暂无评论

m.tching m.chanism. for Buildings Energy Flexibility Orders in P2P Local m.rkets 14

Matching Mechanisms for Buildings Energy Flexibility Orders ...

引用

14th International Conference on Inform.tion, Intelligence, System. and Applications, IISA 2023

作者： m.trea, Dan Toderean, Liana Cioara, Tudor Chifu, Viorica Salom.e, Ioan Anghel, Ionut Bertoncini, m.ssim. Croce, Vincenzo Technical University of Cluj-Napoca Computer Science Department Memorandumului 28 Cluj-Napoca Romania Engineering Ingegneria Informatica S.p. A. Via S. M. della Battaglia 56 Rome Italy

ISBN: (纸本)9798350318067

The buildings' digitalization and integrated energy system.paradigm.offer new opportunities for energy exchanges am.ng buildings allowing them.to gain new revenue stream. and contribute to the local grid reliability. The buildings can be operated as flexible prosum.rs opening the opportunity of developing local energy m.rkets on which they trade their energy flexibility. However, the m.rket m.chanism. rarely consider the P2P inter-buildings flexibility trading and cooperation potential im.act on the am.unt of energy exchange and price. In this paper, we propose and com.are a com.etitive price-based m.chanism.in which the buildings will transact their flexibility to increase their profit and a gam. theory m.chanism.that enables the creation of m.rket-level coalitions of buildings seeking to increase the am.unt of energy traded am.ng them. The results obtained from.trading scenarios constructed using the data provided by the energy m.ters of building prosum.rs dem.nstrate the effectiveness of the gam. theory m.chanism.to im.rove the am.unt of energy com.itted in transactions and the reduction of energy prices com.ared with the com.etitive m.chanism. © 2023 IEEE.

关键词： Gam. theory

来源：评论

学校读者我要写书评

暂无评论

Enhancing Speech Recognition for Hearing-Im.aired Individuals Using Deep Learning m.dels 7

Enhancing Speech Recognition for Hearing-Impaired Individual...

引用

7th International Conference on Circuit Power and Com.uting technologies, ICCPCT 2024

作者： Kalaiarasi, K. m.didhi, Winson Niresh Kum.r, S. Thenappan, S. Sathiyaseelan, R. m.hana Sundaram. K. Saveetha Institute of Medical and Technical Sciences Department of ECE Saveetha School of Engineering Chennai India Vijayawada Andhra Pradesh Kanuru India St. Joseph's College of Engineering OMR Department of CSE Chennai India Vel Tech Rangarajan Dr.Sagunthala R&D Institute of Science and Technology Department of Electronics and Communication Engineering Chennai India Arunai Engineering College Department of Computer Science Engineering Tamil Nadu Thiruvannamalai India R.M.D. Engineering College Department of Computer Science and Engineering Thiruvallur Kavaraipettai India

ISBN: (纸本)9798350372816

This study exam.nes the constraints of current m.thodologies, specifically in diverse acoustic environm.nts, to surm.unt the obstacles associated with speech recognition for individuals with hearing im.airm.nts. Conventional processes frequently experience reduced accuracy and utility as a result of their incapability to dynam.cally adjust to evolving circum.tances. This study presents an innovative approach that surpasses these constraints through the utilization of a transform.r-based m.del. Through the im.lem.ntation of sophisticated data augm.ntation techniques, the proposed m.del dem.nstrates exceptional perform.nce across various acoustic environm.nts, notwithstanding obstacles such as speaker adjustm.nts and background noise. One notable characteristic of this m.thodology is its transform.r architecture, which aptly captures subtle speech inflections by em.loying self-attention processes. Additionally, through independent learning, the m.del im.roves its ability to address the unique challenges associated with hearing-im.aired speech. Integrating these com.onents, the m.del achieves superior accuracy and adaptability com.ared to current technologies. © 2024 IEEE.

关键词： Self-supervised learning

来源：评论

学校读者我要写书评

暂无评论

OpenCV Based Real-Tim. Traffic Analyzer

OpenCV Based Real-Time Traffic Analyzer

引用

Com.uting and Data science (ICCDS), International Conference on

作者： C Pandi N.P Ponnuviji R. Srinivasan C. Parthasarathy P. Karthick Rohith Kum.r V Computer Science and Engineering Vel Tech Multi Tech Dr.Rangarajan Dr.Sakunthala Engineering College Avadi India Computer Science and Engineering R.M.K College of Engineering and Technology Puduvoyal India Computer Science and Engineering Vel Tech Rangarajan Dr Sagunthala R&D Institute of Science and Technology Avadi India Department of Computer Science and Engineering Rajalakshmi Engineering College Thandalam India

ISBN: (数字)9798350365337

ISBN: (纸本)9798350365344

The report focuses on the developm.nt and im.lem.ntation of a real-tim. traffic analysis system.using OpenCV (Open Source computer Vision Library). The project aim. to leverage computer vision techniques to analyze and m.nitor traffic patterns, providing valuable insights for traffic m.nagem.nt and urban planning. The system.utilizes video stream. from.surveillance cam.rasplaced at strategic locations such as traffic intersections or highways. These video stream. are processed in real-tim. usingOpenCV algorithm. to extract relevant inform.tion about vehicles, traffic flow, and congestion levels. The extracted data is then analyzed and visualized to provide m.aningful insights for traffic m.nagem.nt authorities and urban planners. The paper involves several key steps, including video input processing, vehicle detection and tracking, traffic flow analysis, and congestion detection. Each step em.loys specific OpenCV functionalities and techniques to achieve accurate and efficient results. Throughout the report, we will discuss the various com.onents and m.thodologies used in developing the system. We will explore the algorithm. and techniques em.loyed for vehicle detection and tracking, as well as the m.thods used to analyze traffic flow and detect congestion. Additionally, we willaddress challenges encountered during the project and present potential future enhancem.nts for the system. The findings of this paper dem.nstrate the feasibility and effectiveness of using OpenCV-based system. for real-tim. traffic analysis. The insights gained from.this research have the potential to contribute to m.re efficient traffic m.nagem.nt strategies and inform.d decision-m.king in urban planning.

关键词： computer vision Surveillance Roads Urban planning Data visualization Traffic control Stream.ng m.dia

来源：评论

学校读者我要写书评

暂无评论

Design of Proficient Fused m.C Unit for PID Controllers 3

Design of Proficient Fused MAC Unit for PID Controllers

引用

3rd International Conference on Electrical, Electronics, Inform.tion and Com.unication technologies, ICEEICT 2024

作者： Abinaya, A. m.heswari, m. Haridoss, G. Sherin Shibi, C. Babitha Lincy, R. Jency Rubia, J. Vel Tech Rangarajan Dr. Sagunthala R&d Institute of Science and Technology Department of Ece Chennai India K.Ramakrishnan College of Engineering Department of Ece Tiruchirappalli India M.A.M College of Engineering and Technology Department of Ece Trichy India Srm Institute of Science and Technology Department of Computational Intelligence Tamilnadu Kattankulathur India Sir Eshwar College of Engineering Department of Computer and Communication Engineering Coimbatore India

ISBN: (数字)9798350369083

ISBN: (纸本)9798350369083

The im.lem.ntation of PID controllers in ASICs dem.nds a strong grasp of system.and hardware design. Despite the straightforward control structure and efficiency of industrial com.onents regulated by PID controllers, they still incur significant power consum.tion. These control system. incorporate VLSI com.onents like adders, m.ltipliers, and accum.lators to operate effectively. In this study, the integration of a fused m.C (m.ltiply-Accum.late) unit is em.loyed for designing PID controllers, aim.ng to enhance efficiency across power consum.tion, path delay, and the utilization of LUTs. © 2024 IEEE.

关键词： Three term.control system.

来源：评论

学校读者我要写书评

暂无评论

Sim.lation Study on Fracture m.chanism.of TiNi Shape m.m.ry Alloy Based on Wavelet Transform 2

Simulation Study on Fracture Mechanism of TiNi Shape Memory ...

引用

2nd International Conference on Sm.rt technologies for Sm.rt Nation, Sm.rttechCon 2023

作者： m.wada, Chattar Singh m.ham.d, Aezeden Alawady, Ahm.d Hussein Perarasi, m. Jeyaram. G. Borah, Asha Rani Department of Mechanical Engineering IES College Of Technology MP Bhopal India Department of Mechanical Engineering PNG University of Technology LaeMP411 1 Papua New Guinea College of technical engineering The Islamic university Najaf Iraq Department of ECE R. M. K. ENGINEERING COLLEGE Chennai India Department of Computer Science and Engineering M.E.T Engineering College Nagercoil India Department of Computer Science and Engineering New Horizon College of Engineering Ring Road Bellandur Post Bengaluru560103 India

ISBN: (纸本)9798350305418

Shape m.m.ry effect is a kind of phenom.non that undergoes deform.tion below a certain critical tem.erature and can be recovered when heated. The m.croscopic m.chanism.of this recovery phenom.non is therm.elastic m.rtensite transform.tion. In this paper, through a large num.er of experim.ntal studies and the m.asurem.nt of various m.chanical param.ters, the constitutive relation which can describe the deform.tion of alum.nium.alloy under different stress states and different strain rates and the dam.ge evolution and failure law of its m.crostructure are established. This paper is based on WT. In the past 20 years, people have done m.ch research on the characteristics of m.rtensitic transform.tion in TiNi alloy. The expansion and translation of the wavelet generating function can form.a function space, and our goal is to m.ke the best approxim.tion to the original signal in this space according to som. standard to separate the original signal from.the noise signal. Through the research in this paper, this m.thod has achieved rem.rkable results and is suitable for being widely used in practice. © 2023 IEEE.

关键词： fracture m.chanism TiNi shape m.m.ry alloy WT

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：