检索结果-内蒙古大学图书馆

Hybrid Machine Learning Algorithms for Optimal Diagnosis of Heart Disease with Feature Analysis

学校读者我要写书评

暂无评论

Hybrid Machine Learning Algorithms for Optimal Diagnosis of ...

Power, Instrumentation, Energy and Control (PIECON), International Conference on

作者： Ghulab Nabi Ahmad Hira Fatima Shafiullah Mohammad Haris Institute of Applied Sciences Mangalayatan University Aligarh India Department of Mathematics K.C.T.C College Raxual BRA Bihar University Muzaffarpur India Department of Computer Science and Engineering Institute of Technology and Management Aligarh India

Timely prediction of heart disease and its cause is the most challenging issue in medical science. This paper uses various machine learning algorithms such as Logistic Regression, Support Vector Machine, Decision Tree, Random Forest, Naive Bayes, K-Nearest Neighbourhood, and K-fold cross-validation are used to predict heart diseases. The system uses a K-fold cross-validation technique to enhance the accuracies of algorithms. The UCI Kaggle Cleveland heart disease datasets is used to analyse the performance of the models. It is found in the experiment that the training accuracy of K-Nearest Neighbour is 88.52%, and Recall is 93.30%. The Random Forest produced the highest and most comparable Receiver Operating Characteristics Curve accuracy. Moreover, the experimental results of the recommended techniques are compared with previous heart disease prediction studies, and it is found that among the suggested technique, the performance of K-Nearest Neighbour is best. The fundamental goal of this study is to design a novel and distinctive model-creation approach for resolving practical issues.

关键词： Heart Training Support vector machines Machine learning algorithms Instruments Receivers Prediction algorithms

A Comparative Study of Data Augmentation Techniques for Deep Learning Based Emotion Recognition

学校读者我要写书评

暂无评论

arXiv 2022年

作者： Shankar, Ravi Kenfack, Abdouh Harouna Somayazulu, Arjun Venkataraman, Archana Department of Electrical and Computer Engineering Johns Hopkins University United States Department of Applied Mathematics and Statistics Johns Hopkins University United States Department of Computer Science Johns Hopkins University United States

Automated emotion recognition in speech is a long-standing problem. While early work on emotion recognition relied on hand-crafted features and simple classifiers, the field has now embraced end-to-end feature learning and classification using deep neural networks. In parallel to these models, researchers have proposed several data augmentation techniques to increase the size and variability of existing labeled datasets. Despite many seminal contributions in the field, we still have a poor understanding of the interplay between the network architecture and the choice of data augmentation. Moreover, only a handful of studies demonstrate the generalizability of a particular model across multiple datasets, which is a prerequisite for robust real-world performance. In this paper, we conduct a comprehensive evaluation of popular deep learning approaches for emotion recognition. To eliminate bias, we fix the model architectures and optimization hyperparameters using the VESUS dataset and then use repeated 5-fold cross validation to evaluate the performance on the IEMOCAP and CREMA-D datasets. Our results demonstrate that long-range dependencies in the speech signal are critical for emotion recognition and that speed/rate augmentation offers the most robust performance gain across models. © 2022, CC BY.

关键词： Emotion Recognition

Classifying Activities of Electrical Line Workers Based on Deep Learning Approaches Using Wrist-Worn Sensor

学校读者我要写书评

暂无评论

Classifying Activities of Electrical Line Workers Based on D...

International Joint Conference on computer science and Software engineering (JCSSE)

作者： Sakorn Mekruksavanich Anuchit Jitpattanakul Department of Computer Engineering School of Information and Communication Technology University of Phayao Thailand Department of Mathematics Intelligent and Nonlinear Dynamic Innovations Research Center Faculty of Applied Science King Mongkut's University of Technology North Bangkok Bangkok Thailand

The field o f human activity recognition (HAR) is a priority for cutting-edge study because of its potential to revolutionize the way we understand and improve our everyday lives. A large variety of ordinary, everyday tasks has been classified using H AR. Nevertheless, inc ontrast to basic human actions, the increasing demands of numerous real-world applications have attracted the interest of the HAR area of study. Electrical line workers (ELWs) face a variety of challenges, including long hours, working in isolated locations, and performing particularly hazardous tasks. Wearable sensor-based HAR allows for unobtrusive tracking of ELW efficiency and security. This study explores deep learning strategies for automatically categorizing ELWs' complicated actions through sensor data collected through a wrist-worn device. We propose ResNeXt, a deep residual neural network, and evaluate it with other deep learning networks for their ability to categorize ELW activities effectively. We employ a publicly available benchmark dataset that includes 10 ELW tasks. The results of the experiment demonstrate that the proposed ResNeXt achieved the highest accuracy (98.74%) and F1-score (98.81%) compared to other deep learning networks studied.

关键词：

MetaUE: Model-based Meta-learning for Underwater Image Enhancement

学校读者我要写书评

暂无评论

arXiv 2023年

作者： Zhang, Zhenwei Yan, Haorui Tang, Ke Duan, Yuping Center for Applied Mathematics Tianjin University Tianjin300072 China The Guangdong Key Laboratory of Brain-Inspired Intelligent Computation Department of Computer Science and Engineering Southern University of Science and Technology Shenzhen518055 China The Research Institute of Trustworthy Autonomous Systems Southern University of Science and Technology Shenzhen518055 China

The challenges in recovering underwater images are the presence of diverse degradation factors and the lack of ground truth images. Although synthetic underwater image pairs can be used to overcome the problem of inadequately observing data, it may result in over-fitting and enhancement degradation. This paper proposes a model-based deep learning method for restoring clean images under various underwater scenarios, which exhibits good interpretability and generalization ability. More specifically, we build up a multi-variable convolutional neural network model to estimate the clean image, background light and transmission map, respectively. An efficient loss function is also designed to closely integrate the variables based on the underwater image model. The meta-learning strategy is used to obtain a pre-trained model on the synthetic underwater dataset, which contains different types of degradation to cover the various underwater environments. The pre-trained model is then fine-tuned on real underwater datasets to obtain a reliable underwater image enhancement model, called MetaUE. Numerical experiments demonstrate that the pre-trained model has good generalization ability, allowing it to remove the color degradation for various underwater attenuation images such as blue, green and yellow, etc. The fine-tuning makes the model able to adapt to different underwater datasets, the enhancement results of which outperform the state-of-the-art underwater image restoration methods. All our codes and data are available at https://***/Duanlab123/MetaUE. Copyright © 2023, The Authors. All rights reserved.

关键词： Image enhancement

Resilient Distributed Optimization

学校读者我要写书评

暂无评论

arXiv 2022年

作者： Zhu, Jingxuan Lin, Yixuan Velasquez, Alvaro Liu, Ji The Department of Applied Mathematics and Statistics Stony Brook University United States The Department of Computer Science University of Colorado Boulder United States The Department of Electrical and Computer Engineering Stony Brook University United States

This paper considers a distributed optimization problem in the presence of Byzantine agents capable of introducing untrustworthy information into the communication network. A resilient distributed subgradient algorithm is proposed based on graph redundancy and objective redundancy. It is shown that the algorithm causes all non-Byzantine agents' states to asymptotically converge to the same optimal point under appropriate assumptions. A partial convergence rate result is also provided. Copyright © 2022, The Authors. All rights reserved.

关键词： Redundancy

Utilizing probabilistic entanglement between sensors in quantum networks

学校读者我要写书评

暂无评论

Physical Review applied 2024年第6期22卷 064085-064085页

作者： Emily A. Van Milligen Christos N. Gagatsos Eneet Kaur Don Towsley Saikat Guha Department of Physics The University of Arizona 1630 East University Boulevard Tucson Arizona 85721 USA Department of Electrical and Computer Engineering The University of Arizona 1630 East University Boulevard Tucson Arizona 85721 USA Wyant College of Optical Sciences The University of Arizona 1630 East University Boulevard Tucson Arizona 85721 USA Program in Applied Mathematics The University of Arizona Tucson Arizona 85721 USA Cisco Quantum Lab Los Angeles USA College of Computer Science University of Massachusetts Amherst Massachusetts USA Department of Electrical and Computer Engineering The University of Maryland College Park Maryland 20742 USA

One of the most promising applications of quantum networks is entanglement-assisted sensing. The field of quantum metrology exploits quantum correlations to improve the precision bound for applications such as precision timekeeping, field sensing, and biological imaging. When measuring multiple spatially distributed parameters, current literature focuses on quantum entanglement in the discrete-variable case and quantum squeezing in the continuous-variable case, distributed amongst all of the sensors in a given network. However, it can be difficult to ensure that all sensors preshare entanglement of sufficiently high fidelity. This work probes the space between fully entangled and fully classical sensing networks by modeling a star network with probabilistic entanglement generation that is attempting to estimate the average of local parameters. The quantum Fisher information is used to determine which protocols best utilize entanglement as a resource for different network conditions. It is shown that without entanglement distillation there is a threshold fidelity below which classical sensing is preferable. For a network with a given number of sensors and links characterized by a certain initial fidelity and probability of success, this work outlines when and how to use entanglement, when to store it, and when it needs to be distilled.

关键词： Quantum entanglement

Wrinkling instability of 3D auxetic bilayers in tension

学校读者我要写书评

暂无评论

arXiv 2024年

作者： Venkata, Sairam Pamulaparthi Fu, Yuxin Fu, Yibin Danesh, Hooman Destrade, Michel Balbi, Valentina School of Mathematical and Statistical Sciences University of Galway Galway Ireland Department of Mechanics Tianjin University Tianjin China School of Computer Science and Mathematics Keele University United Kingdom Institute of Applied Mechanics RWTH Aachen Aachen Germany Department of Engineering Mechanics Zhejiang University Hangzhou China

Bilayers, soft substrates coated with stiff films, are commonly found in nature with examples including skin tissue, vesicles, and organ membranes. They exhibit different types of instabilities when subjected to compression, depending on the contrast in material properties between the two components. In this work, we unravel the mechanisms behind wrinkling instabilities in auxetic bilayer systems under uniaxial tension. We find that a soft bilayer in tension can experience significant lateral contraction, and with sufficient contrast in Poisson ratios, compressive stresses may induce wrinkles aligned with the tensile direction. We analytically model the onset of wrinkles and validate our predictions using Finite Element simulations in ABAQUS. Our findings reveal that wrinkles occur when the Poisson ratio of the substrate is greater than that of the film. As the two Poisson ratios converge to a common value, the critical stretch for instability shoots up rapidly and the wrinkles disappear. We also confirm these results through asymptotic analysis. Using inverse analysis, we design film microstructures to achieve desired effective Poisson ratios and further validate the effective properties with the Finite Element code FEAP. We show that the critical stretch ratio for buckling in auxetic structures with microstructural patterns is in strong agreement with the homogenized model predictions. The proposed method has significant potential for controlling surface patterns in auxetic skin grafts and hydrogel organ patches under mechanical loads. Moreover, the asymptotic expressions for compressible bilayers developed in this work can also be applied under finite strain for buckling-based metrology. © 2024, CC BY.

关键词： Poisson ratio

Simulation of Conducted Emissions From Power Converters Using Leading Switching Technologies

学校读者我要写书评

暂无评论

Simulation of Conducted Emissions From Power Converters Usin...

IEEE Global Electromagnetic Compatibility Conference (GEMCCON)

作者： Amr Madi Francinei Vieira Piotr Lezynski Sebastian Koj Robert Smolenski Electrical Engineering Mathematics & Computer Science (EEMCS) University of Twente Enschede Netherlands Institute of Electrical Engineering and Measurement Technology Leibniz University Hannover Hannover Germany Institute of Automatic Control Electronics and Electrical Engineering University of Zielona Gora Zielona Gora Poland Department of Engineering Sciences Jade University of Applied Sciences Wilhelmshaven Germany

This paper presents a comparative study between the commonly used switches (IGBT, MOSFET, SiC and GaN) for power inverters of electric vehicles. A 400 V, three-phase inverter system simulation test is implemented using LTspice and the conducted emissions are compared to variant CISPR standards and switching frequencies in order to show the possible interferences with the other car equipment.

关键词： MOSFET Silicon carbide Switching frequency Interference Predictive models SPICE Inverters

Deep Learning Approaches for Recognizing Daily Human Activities Using Smart Home Sensors

学校读者我要写书评

暂无评论

Deep Learning Approaches for Recognizing Daily Human Activit...

International ECTI Northern Section Conference on Electrical, Electronics, computer and Telecommunications engineering (ECTI-NCON)

作者： Worrakit Sanpote Ponnipa Jantawong Narit Hnoohom Anuchit Jitpattanakul Sakorn Mekruksavanich Department of Computer Science School of Information and Communication Technology University of Phayao Phayao Thailand Department of Computer Engineering School of Information and Communication Technology University of Phayao Phayao Thailand Department of Computer Engineering Image Information and Intelligence Laboratory Faculty of Engineering Mahidol University Nakorn Pathom Thailand Department of Mathematics Intelligent and Nonlinear Dynamic Innovations Research Center Faculty of Applied Science King Mongkut's University of Technology North Bangkok Bangkok Thailand

Nowadays, one of the most important objectives in health-related research is the improvement of the living condition and well-being of people. Smart home systems can provide health protection for residents based on the results of daily activity recognition. Recent advances and developments in sensor technology have increased the need for sensor-compatible goods and services in smart homes. Consequently, the ever-increasing volume of data requires the field of deep learning (DL) for auto-matic human motion recognition. Recent research has modeled spatiotemporal sequences gathered by smart home sensors using long short-term memory networks. In this work, ResNeXt-based models that learn to classify human activities in smart homes were proposed to improve recognition performance. Experiments conducted on Center for Advanced Studies in Adaptive Systems (CASAS) data, a publicly available benchmark dataset, shows that the proposed ResNeXt-based techniques are significantly superior to the existing DL methods and provide better results compared to the existing literature. The ResNeXt model achieved the averaged accuracy over the benchmark method to 84.81%, 93.57%, and 90.38% for the CASAS_Cairo, CASAS_Milan and CASAS_Kyoto3 datasets, respectively.

关键词：