检索结果-南通市图书馆

Machine Fault Diagnosis Using Audio Sensors data and Explainable ai Techniques-LIME and SHAP

在线全文

维普期刊数据库

学校读者我要写书评

暂无评论

Computers, Materials and Continua 2024年第3期80卷 3463-3484页

作者： Zereen A.N. Das A. Uddin J. School of data and Sciences Brac University Dhaka 1212 Bangladesh JW KIM College of Future Studies Endicott College Woosong University Daejeon 300-718 South Korea ai and big data department Endicott College Woosong University Daejeon 300-718 South Korea

Machine fault diagnostics are essential for industrial operations, and advancements in machine learning have significantly advanced these systems by providing accurate predictions and expedited solutions. Machine learning models, especially those utilizing complex algorithms like deep learning, have demonstrated major potential in extracting important information from large operational datasets. Despite their efficiency, machine learning models face challenges, making Explainable ai (Xai) crucial for improving their understandability and fine-tuning. The importance of feature contribution and selection using Xai in the diagnosis of machine faults is examined in this study. The technique is applied to evaluate different machine-learning algorithms. Extreme Gradient Boosting, Support Vector Machine, Gaussian Naive Bayes, and Random Forest classifiers are used alongside Logistic Regression (LR) as a baseline model because their efficacy and simplicity are evaluated thoroughly with empirical analysis. The Xai is used as a targeted feature selection technique to select among 29 features of the time and frequency domain. The Xai approach is lightweight, trained with only targeted features, and achieved similar results as the traditional approach. The accuracy without Xai on baseline LR is 79.57%, whereas the approach with Xai on LR is 80.28%. Copyright © 2024 The Authors. Published by Tech Science Press.

关键词： Explainable ai feature selection machine fault diagnosis machine learning

A Distributed Computing Algorithm for Electricity Carbon Emission Flow and Carbon Emission Intensity

在线全文

维普期刊数据库

学校读者我要写书评

暂无评论

Protection and Control of Modern Power Systems 2024年第2期9卷 138-146页

作者： Xingping Wu Wei Yang Ning Zhang Chunlei Zhou Jinwei Song Chongqing Kang big data Center of China Network Chinese Society of Electrical Engineering State Network big data Center department of Electrical Engineering Tsinghua UniversityBeijingChina 不详 big data Center State Grid Corporation of ChinaBeijing 100052China department of Electrical Engi-neering Tsinghua UniversityBeijingChina

The calculation of the indirect carbon emis-sion is essential for power system policy making,carbon market development,and power grid *** em-bedded carbon emissions of the electricity system are commonly calculated by carbon emission flow ***,the calculation procedure is time-consuming,especially for a country with 500-1000 thousand nodes,making it challenging to obtain nationwide carbon emis-sions intensity ***,the calculation procedure requires to gather all the grid data with high classified levels from different power grid companies,which can prevent data sharing and cooperation among different *** paper proposes a distributed computing algorithm for indirect carbon emission that can reduce the time consumption and provide privacy *** core idea is to utilize the sparsity of the nodes’flow matrix of the nationwide grid to partition the computing procedure into parallel sub-procedures exe-cuted in multiple *** flow and structure data of the regional grid are transformed irreversibly for pri-vacy protection,when transmitted between terminals.A 1-master-and-N-slave layout is adopted to verify the *** algorithm is suitable for large grid compa-nies with headquarter and branches in provinces,such as the State Grid Corporation of China.

关键词： Carbon emission flow cooperative computing carbon emission intensity matrix block par-tition power flow tracing parallel computing privacy protection

Correlation analysis between lung ultrasound scores and pulmonary arterial systolic pressure in patients with acute heart failure admitted to the emergency intensive care unit

在线全文

维普期刊数据库

学校读者我要写书评

暂无评论

Journal of Intensive Medicine 2024年第1期4卷 125-132页

作者： Ping Xu Basma Nasr Liang Li Wenbin Huang Wei Liu Xuelian Wang Emergency department Zigong Fourth People’s HospitalZigongChina Institute of Medical big data Zigong Academy of Artificial Intelligence and Big Data for Medical ScienceZigongChina Artificial Intelligence Key Laboratory of Sichuan Province ZigongChina department of Cardiology First Affiliated Hospital of Dalian Medical UniversityDalianChina

Background:No convenient,inexpensive,and non-invasive screening tools exist to identify pulmonary hypertension(PH)-left heart disease(LHD)patients during the early stages of the disease *** study investigated whether different methods of lung ultrasound(LUS)could be used for the initial investigation of ***:This was a single-center prospective observational study which was performed in the Zigong Fourth People’s *** consecutively enrolled patients with heart failure(HF)admitted to the emergency intensive care unit from January 2018 to May *** echocardiography and LUS were performed within 24 h before *** used the Spearman coefficient for correlation analysis between ultrasound scores and pulmonary arterial systolic pressure(PASP).Bland-Altman plots were generated to inspect possible bias,and receiver operating characteristic(ROC)curves were calculated to assess the relationship between ultrasound scores and an intermediate and high echocardiographic probability of ***:Seventy-one patients were enrolled in this study,with an overall median age of 79(interquartile range:71.5–84.0)*** the 71 patients,36(50.7%)cases were male,and 26(36.6%)had an intermediate and high echocardiographic probability of *** four LUS scores in patients with an intermediate and high probability of PH were significantly higher than in patients with a low probability of PH(P<0.05).The correlation coefficient(r)between different LUS scoring methods and PASP was moderate for the 6-zone(r=0.455,P<0.001),8-zone(r=0.385,P=0.001),12-zone(r=0.587,P<0.001),and 28-zone(r=0.535,P<0.001)*** Bland-Altman plots,each of the four LUS scoring methods had a good agreement with PASP(P<0.001).The 8-zone and 12-zone methods showed moderately accurate discriminative values in differentiating patients with an intermediate and high echocardiographic probability of PH(P<0.05).

关键词： Acute heart failure Pulmonary arterial systolic pressure Pulmonary hypertension Left heart disease Lung ultrasound

Two kinds of average approximation accuracy

在线全文

维普期刊数据库

学校读者我要写书评

暂无评论

CAai Transactions on Intelligence Technology 2024年第2期9卷 481-490页

作者： Qingzhao Kong Wanting Wang Dongxiao Zhang Wenbin Zhang department of Science Jimei UniversityXiamenChina Digital Fujian big data Modeling and Intelligent Computing Institute Jimei UniversityXiamenChina department of Computer Science Michigan Technological UniversityHoughtonMichiganUSA

Rough set theory places great importance on approximation accuracy,which is used to gauge how well a rough set model describes a target ***,traditional approximation accuracy has limitations since it varies with changes in the target concept and cannot evaluate the overall descriptive ability of a rough set *** overcome this,two types of average approximation accuracy that objectively assess a rough set model’s ability to approximate all information granules is *** first is the relative average approximation accuracy,which is based on all sets in the universe and has several basic *** second is the absolute average approximation accuracy,which is based on undefinable sets and has yielded significant *** also explore the relationship between these two types of average approximation ***,the average approximation accuracy has practical applications in addressing missing attribute values in incomplete information tables.

关键词： rough sets rough set theory

Temporal Second-order Scheme for a Hidden-memory Variable Order Time Fractional Diffusion Equation with an Initial Singularity

在线全文

维普期刊数据库

学校读者我要写书评

暂无评论

Acta Mathematicae Applicatae Sinica 2024年第4期 1060-1077页

作者： Rui-lian DU Zhi-zhong SUN School of big data Changzhou University department of Mathematics Shanghai University School of Mathematics Southeast University

In this work,a novel time-stepping ■ formula is developed for a hidden-memory variable-order Caputo’s fractional derivative with an initial *** formula can obtain second-order accuracy and an error estimate is analyzed *** an application,a fully discrete difference scheme is established for the initial-boundary value problem of a hidden-memory variable-order time fractional diffusion *** experiments are provided to support our theoretical results.

关键词：

Influencing Factors of Healthy Aging Risk Assessed Using Biomarkers:A Life Course Perspective

同方期刊数据库评论

在线全文

同方期刊数据库

学校读者我要写书评

暂无评论

China CDC weekly 2024年第11期6卷 219-224,I0014-I0017页

作者： Cedric Zhang Bo Lua Yajie Gao Jinming Li Xingqi Cao Xinwei Lyu Yinuo Tu Shuyi Jin Zuyun Liu Center for Clinical big data and Analytics of the Second Affiliated Hospital and Department of Big Data in Health Science School of Public Healththe Key Laboratory of Intelligent Preventive Medicine of Zhejiang ProvinceZhejiang University School of MedicineHangzhou CityZhejiang ProvinceChina Institute of Epidemiology and Health Care University College LondonLondonUK College of Chemical and Biological Engineering Zhejiang UniversityHangzhou CityZhejiang ProvinceChina

Assessing individual risks of healthy aging using biomarkers and identifying associated factors have become important areas of *** this study,we conducted a literature review of relevant publications between 2018 and 2023 in both Chinese and English *** studies have predominantly used single biomarkers,such as C-reactive protein,or focused on specific life course stages and factors such as socioeconomic status,mental health,educational levels,and unhealthy *** summarizing the progress in this field,our study provides valuable insights and future directions for promoting healthy aging from a life course perspective.

关键词： such insight directions

Classification of Sailboat Tell Tail Based on Deep Learning

在线全文

维普期刊数据库

学校读者我要写书评

暂无评论

Journal of Ocean University of China 2024年第3期23卷 710-720页

作者： CHANG Xiaofeng YU Jintao GAO Ying DING Hongchen LIU Yulong YU Huaming department of Physical Education Ocean University of ChinaQingdao 266100China College of big data Qingdao University of Science and TechnologyQingdao 266061China College of Oceanic and Atmospheric Sciences Ocean University of ChinaQingdao 266100China

The tell tail is usually placed on the triangular sail to display the running state of the air flow on the sail *** is of great significance to make accurate judgement on the drift of the tell tail of the sailboat during sailing for the best sailing *** it is difficult for sailors to keep an eye for a long time on the tell sail for accurate judging its changes,affected by strong sunlight and visual *** this case,we adopt computer vision technology in hope of helping the sailors judge the changes of the tell tail in ease with *** paper proposes for the first time a method to classify sailboat tell tails based on deep learning and an expert guidance system,supported by a sailboat tell tail classification data set on the expert guidance system of interpreting the tell tails states in different sea wind conditions,including the feature extraction *** the expression capabilities that vary with the computational features in different visual tasks,the paper focuses on five tell tail computing features,which are recoded by an automatic encoder and classified by a SVM *** experimental samples were randomly divided into five groups,and four groups were selected from each group as the training set to train the *** remaining one group was used as the test set for *** highest resolution value of the ResNet network was 80.26%.To achieve better operational results on the basis of deep computing features obtained through the ResNet network in the *** method can be used to assist the sailors in making better judgement about the tell tail changes during sailing.

关键词： tell tail sailboat classification deep learning

IMTNet: Improved Multi-Task Copy-Move Forgery Detection Network with Feature Decoupling and Multi-Feature Pyramid

维普期刊数据库

同方期刊数据库评论

在线全文

学校读者我要写书评

暂无评论

Computers, Materials and Continua 2024年第3期80卷 4603-4620页

作者： Wang H. Jiang Z. Qian Q. Long Y. School of Information Guizhou University of Finance and Economics Guiyang 550025 China College of big data Statistics Guizhou University of Finance and Economics Guiyang 550025 China

Copy-Move Forgery Detection (CMFD) is a technique that is designed to identify image tampering and locate suspicious areas. However, the practicality of the CMFD is impeded by the scarcity of datasets, inadequate quality and quantity, and a narrow range of applicable tasks. These limitations significantly restrict the capacity and applicability of CMFD. To overcome the limitations of existing methods, a novel solution called IMTNet is proposed for CMFD by employing a feature decoupling approach. Firstly, this study formulates the objective task and network relationship as an optimization problem using transfer learning. Furthermore, it thoroughly discusses and analyzes the relationship between CMFD and deep network architecture by employing ResNet-50 during the optimization solving phase. Secondly, a quantitative comparison between fine-tuning and feature decoupling is conducted to evaluate the degree of similarity between the image classification and CMFD domains by the enhanced ResNet-50. Finally, suspicious regions are localized using a feature pyramid network with bottom-up path augmentation. Experimental results demonstrate that IMTNet achieves faster convergence, shorter training times, and favorable generalization performance compared to existing methods. Moreover, it is shown that IMTNet significantly outperforms fine-tuning based approaches in terms of accuracy and F1 Copyright © 2024 The Authors. Published by Tech Science Press.

关键词： feature decoupling Image copy-move detection multi-scale feature pyramids passive forensics

Sparse representation scheme with enhanced medium pixel intensity for face recognition

在线全文

维普期刊数据库

学校读者我要写书评

暂无评论

CAai Transactions on Intelligence Technology 2024年第1期9卷 116-127页

作者： Xuexue Zhang Yongjun Zhang Zewei Wang Wei Long Weihao Gao Bob Zhang State Key Laboratory of Public big data Institute for Artificial IntelligenceCollege of Computer Science and TechnologyGuizhou UniversityGuiyangGuizhouChina department of Computer and Information Science University of MacaoAvenida da UniversidadeTaipaMacaoChina

Sparse representation is an effective data classification algorithm that depends on the known training samples to categorise the test *** has been widely used in various image classification *** in sparse representation means that only a few of instances selected from all training samples can effectively convey the essential class-specific information of the test sample,which is very important for *** deformable images such as human faces,pixels at the same location of different images of the same subject usually have different ***,extracting features and correctly classifying such deformable objects is very ***,the lighting,attitude and occlusion cause more *** the problems and challenges listed above,a novel image representation and classification algorithm is ***,the authors’algorithm generates virtual samples by a non-linear variation *** method can effectively extract the low-frequency information of space-domain features of the original image,which is very useful for representing deformable *** combination of the original and virtual samples is more beneficial to improve the clas-sification performance and robustness of the ***,the authors’algorithm calculates the expression coefficients of the original and virtual samples separately using the sparse representation principle and obtains the final score by a designed efficient score fusion *** weighting coefficients in the score fusion scheme are set entirely ***,the algorithm classifies the samples based on the final *** experimental results show that our method performs better classification than conventional sparse representation algorithms.

关键词： computer vision face recognition image classification image representation