检索结果-内蒙古大学图书馆

Methods for class-imbalanced learning with support vector machines: a review and an empirical evaluation

Soft Computing 2024年第20期28卷 11873-11894页

作者： Rezvani, Salim Pourpanah, Farhad Lim, Chee Peng Wu, Q. M. Jonathan Department of Computer Science Ryerson University TorontoON Canada Department of Electrical and Computer Engineering University of Windsor WindsorON Canada Institute for Intelligent Systems Research and Innovation Deakin University GeelongVIC Australia

This paper presents a review on methods for class-imbalanced learning with the Support Vector Machine (SVM) and its variants. We first explain the structure of SVM and its variants and discuss their inefficiency in learning with class-imbalanced data sets. We introduce a hierarchical categorization of SVM-based models with respect to class-imbalanced learning. Specifically, we categorize SVM-based models into re-sampling, algorithmic, and fusion methods, and discuss the principles of the representative models in each category. In addition, we conduct a series of empirical evaluations to compare the performances of various representative SVM-based models in each category using benchmark imbalanced data sets, ranging from low to high imbalanced ratios. Our findings reveal that while algorithmic methods are less time-consuming owing to no data pre-processing requirements, fusion methods, which combine both re-sampling and algorithmic approaches, generally perform the best, but with a higher computational load. A discussion on research gaps and future research directions is provided. © The Author(s), under exclusive licence to Springer-Verlag GmbH Germany, part of Springer Nature 2024.

关键词： Support vector machines

来源：评论

学校读者我要写书评

暂无评论

Metaheuristic Secure Clustering Scheme for Energy Harvesting Wireless Sensor Networks

引用

computer Systems science & engineering 2023年第4期45卷 497-512页

作者： S.Nithya Roopa P.Anandababu Sibi Amaran Rajesh Verma Department of Computer Science and Engineering Kumaraguru College of TechnologyCoimbatore641049India Department of Computer and Information Science Faculty of ScienceAnnamalai UniversityChidambaram608002India Department of Computing Technologies SRM Institute of Science and TechnologyChennai603203India Department of Electrical Engineering King Khalid UniversityAbha61411Kingdom of Saudi Arabia

Recently,energy harvesting wireless sensor networks(EHWSN)have increased significant attention among research *** harvesting energy from the neighboring environment,the sensors in EHWSN resolve the energy constraint problem and offers lengthened network *** is one of the proficient ways for accomplishing even improved lifetime in *** clustering process intends to appropriately elect the cluster heads(CHs)and construct *** several models are available in the literature,it is still needed to accomplish energy efficiency and security in *** this view,this study develops a novel Chaotic Rider Optimization Based Clustering Protocol for Secure Energy Harvesting Wireless Sensor Networks(CROC-SEHWSN)*** presented CROC-SEHWSN model aims to accomplish energy efficiency by clustering the node in *** CROC-SEHWSN model is based on the integration of chaotic concepts with traditional rider optimization(RO)***,the CROC-SEHWSN model derives a fitness function(FF)involving seven distinct parameters connected to *** accomplish security,trust factor and link quality metrics are considered in the *** design of RO algorithm for secure clustering process shows the novelty of the *** order to demonstrate the enhanced performance of the CROC-SEHWSN approach,a wide range of simulations are carried out and the outcomes are inspected in distinct *** experimental outcome demonstrated the superior performance of the CROC-SEHWSN technique on the recent approaches with maximum network lifetime of 387.40 and 393.30 s under two scenarios.

关键词： Clustering wireless sensor networks network lifetime energy efficiency metaheuristics energy harvesting rider optimization

来源：评论

学校读者我要写书评

暂无评论

Time delay recursive neural network-based direct adaptive control for a piezo-actuated stage

引用

science China(Technological sciences) 2023年第5期66卷 1397-1407页

作者： WANG YiFan ZHOU MiaoLei SHEN ChuanLiang CAO WenJing HUANG XiaoLiang Department of Control Science and Engineering Jilin UniversityChangchun130022China State Key Laboratory of Automotive Simulation and Control Jilin UniversityChangchun130022China Department of Engineering and Applied Sciences Sophia UniversityTokyo102-8554Japan Department of Electrical Engineering Chalmers University of TechnologyGothenburg41296Sweden

Piezo-actuated stage is a core component in micro-nano manufacturing ***,the inherent nonlinearity,such as rate-dependent hysteresis,in the piezo-actuated stage severely impacts its tracking *** study proposes a direct adaptive control(DAC)method to realize high precision *** proposed controller is designed by a time delay recursive neural *** with those existing DAC methods designed under the general Lipschitz condition,the proposed control method can be easily generalized to the actual systems,which have hysteresis ***,a hopfield neural network(HNN)estimator is proposed to adjust the parameters of the proposed controller ***,a modular model consisting of linear submodel,hysteresis submodel,and lumped uncertainties is established based on the HNN estimator to describe the piezoactuated stage in this ***,the performance of the HNN estimator can be exhibited visually through the modeling *** proposed control method eradicates the adverse effects on the control performance arising from the inaccuracy in establishing the offline model and improves the capability to suppress the influence of hysteresis on the tracking accuracy of piezo-actuated stage in comparison with the conventional DAC *** stability of the control system is ***,a series of comparison experiments with a dual neural networks-based data driven adaptive controller are carried out to demonstrate the superiority of the proposed controller.

关键词： piezo-actuated stage direct adaptive control time delay recursive neural network hopfield neural network estimator

来源：评论

学校读者我要写书评

暂无评论

CPL-SLAM: Centralized Collaborative Multi-Robot Visual-Inertial SLAM Using Point-and-Line Features

引用

IEEE Internet of Things Journal 2025年第12期12卷 21866-21875页

作者： Liu, Xin Wen, Shuhuan Liu, Huaping Richard Yu, F. Yanshan University Engineering Research Center The Ministry of Education for Intelligent Control System and Intelligent Equipment Key Lab of Intelligent Rehabilitation and Neuroregulation in Hebei Province Department of Key Lab of Industrial Computer Control Engineering of Hebei Province Qinhuangdao066004 China Tsinghua University Department of Computer Science and Technology Beijing100084 China Shenzhen University College of Computer Science and Software Engineering China Carleton University Canada

Traditional visual-inertial SLAM (Simultaneous Localization and Mapping) systems predominantly rely on feature point matching from a single robot to realize the robot pose estimation and environment map construction. However, in complex scenarios, these traditional systems struggle with issues such as tracking failures due to illumination changes, rapid movements, and low-texture environments, and they perform poorly in terms of mapping efficiency and global consistency. To address these challenges, we propose a centralized collaborative SLAM system that employs both point and line features for tracking in the robot and map fusion in the cloud. The proposed system leverages the fusion of point and line features across all instances in the process, which allows our method to achieve higher localization accuracy in structured, low-texture scenes. With the aid of classifying gravity-aligned vertical lines and spatial parallel lines, the proposed system can deliver faster and more accurate odometry in complex scenes. Furthermore, we developed intra-robot and inter-robot loop closure detection methods based on point and line features, generating a globally consistent sparse point cloud and structured scene map in the cloud. Our method is able to build richer maps while improving accuracy compared to existing methods. Experimental results on public datasets and in real-world environments show that, compared to existing advanced methods, our approach demonstrates better performance. © 2014 IEEE.

关键词： SLAM robotics

来源：评论

学校读者我要写书评

暂无评论

Comparative analysis of machine learning and ensemble approaches for hepatitis B prediction using data mining with synthetic minority oversampling technique

引用

Health and Technology 2024年第1期14卷 109-118页

作者： Alizargar, Azadeh Chang, Yang-Lang Tan, Tan-Hsu Liu, Tsung-Yu Department of Electrical Engineering College of Electrical Engineering and Computer Science National Taipei University of Technology Taipei 10608 Taiwan Department of Multimedia and Game Science Lunghwa University of Science and Technology Taoyuan 333326 Taiwan

Purpose: Hepatitis B, caused by the Hepatitis B virus (HBV), can harm the liver without noticeable symptoms. Early detection is crucial to prevent transmission and enhance recovery. The main goal is to predict Hepatitis B through cost-effective lab test data, by utilizing machine learning. The primary focus is on evaluating the effectiveness of various algorithms in predicting the disease and their potential to enhance early diagnosis capabilities. Methods: Six distinct algorithms (Support Vector Machine, K-nearest Neighbors, Logistic Regression, decision tree, extreme gradient boosting, random forest) were employed alongside an ensemble model. Analysis involved two rounds: considering all features and key attributes. The Synthetic Minority Oversampling Technique (SMOTE) was employed for data imbalance. Various metrics, including the confusion matrix, precision, recall, F1 score, accuracy, receiver operating characteristics (ROC) curve, area under the curve (AUC), and mean absolute error (MAE), were utilized to assess the efficacy of each predictive technique. The National Health and Nutrition Examination Survey (NHANES) dataset was employed. Results: The experimental results demonstrate that the ensemble model attained the highest accuracy (97%) and AUC (0.997) in comparison to existing models. The analysis revealed that specific crucial features possess substantial predictive significance within this model. Conclusion: The study underscores the potential of the ensemble model as a valuable tool for medical practitioners, leveraging cost-effective and readily obtainable laboratory test data to predict Hepatitis B with remarkable accuracy. By facilitating early diagnosis and intervention, this research presents a promising avenue to enhance patient outcomes in the context of Hepatitis B. © 2023, The Author(s) under exclusive licence to International Union for Physical and engineering sciences in Medicine (IUPESM).

关键词： Early detection Ensemble model Index terms— Hepatitis B Liver damage Machine learning SMOTE

来源：评论

学校读者我要写书评

暂无评论

Aerodynamics-Based Collision-Free control of Connected Drones in Complex Urban Low-Altitude Airspace Using Distributional Reinforcement Learning

引用

IEEE Transactions on Vehicular Technology 2024年第7期73卷 9763-9775页

作者： Liao, Bing-Hao Lee, Chao-Yang Wang, Li-Chun National Yang Ming Chiao Tung University Department of Electrical and Computer Engineering Hsinchu City30010 Taiwan National Yunlin University of Science and Technology Department of Computer Science and Information Engineering Yunlin County64002 Taiwan

Recent growth in the number of drones has made traffic management unworkable, particularly in urban areas. The safe operation and optimized navigation of drone swarms are now growing concerns. In this article, we use distributional reinforcement learning to prevent drone collisions via aerodynamic acceleration and deceleration. Although the instantaneous acceleration or deceleration of drones during an encounter influences drone kinematics, we note that drone kinematics itself can prevent collisions via instantaneous acceleration or deceleration. This article's central question, then, is how to take drone kinematics into account to improve air traffic intersection flow performance. First, the intercommunication between drone swarms is established via an Internet of drones to send and receive the current state of each drone in the swarm and detect potential collisions. In the presence of obstacles, we first use distributional reinforcement learning to find the inevitable change in kinematic height and guide descents and ascents through air traffic intersections. Here, the drone's energy consumption during instantaneous acceleration or deceleration cannot be ignored. A properly designed speed control is crucial to the power management of the drone. Distributional reinforcement learning is used to control drone acceleration and deceleration to avoid collision risk while minimizing energy consumption according to the reference energy recorded by each drone. We flew drones to gather energy consumption data and, to validate our approach, we subsequently conducted simulations in which travel time and energy consumption were measured. We demonstrate that the proposed kinematics-based drone collision avoidance method exhibits excellent performance in terms of travel time and power management. © 1967-2012 IEEE.

关键词： Kinematics

来源：评论

学校读者我要写书评

暂无评论

Text Augmentation-Based Model for Emotion Recognition Using Transformers

引用

computers, Materials & Continua 2023年第9期76卷 3523-3547页

作者： Fida Mohammad Mukhtaj Khan Safdar Nawaz Khan Marwat Naveed Jan Neelam Gohar Muhammad Bilal Amal Al-Rasheed Department of Computer Science The University of HaripurHaripur22620Pakistan Department of Computer Systems Engineering Faculty of Electrical and Computer EngineeringUniversity of Engineering and Technology PeshawarPeshawar25120Pakistan Department of Electronics Engineering Technology University of TechnologyNowshera24100Pakistan Department of Computer Science Shaheed Benazir BhuttoWomen UniversityPeshawar25000Pakistan Department of Information Systems College of Computer and Information SciencesPrincess Nourah bint Abdulrahman UniversityRiyadh11671Saudi Arabia

Emotion Recognition in Conversations(ERC)is fundamental in creating emotionally ***-BasedNetwork(GBN)models have gained popularity in detecting conversational contexts for ERC ***,their limited ability to collect and acquire contextual information hinders their *** propose a Text Augmentation-based computational model for recognizing emotions using transformers(TA-MERT)to address *** proposed model uses the Multimodal Emotion Lines Dataset(MELD),which ensures a balanced representation for recognizing human *** used text augmentation techniques to producemore training data,improving the proposed model’s *** encoders train the deep neural network(DNN)model,especially Bidirectional Encoder(BE)representations that capture both forward and backward contextual *** integration improves the accuracy and robustness of the proposed ***,we present a method for balancing the training dataset by creating enhanced samples from the original *** balancing the dataset across all emotion categories,we can lessen the adverse effects of data imbalance on the accuracy of the proposed *** results on the MELD dataset show that TA-MERT outperforms earlier methods,achieving a weighted F1 score of 62.60%and an accuracy of 64.36%.Overall,the proposed TA-MERT model solves the GBN models’weaknesses in obtaining contextual data for ***-MERT model recognizes human emotions more accurately by employing text augmentation and transformer-based *** balanced dataset and the additional training samples also enhance its *** findings highlight the significance of transformer-based approaches for special emotion recognition in conversations.

关键词： Emotion recognition in conversation graph-based network text augmentation-basedmodel multimodal emotion lines dataset bidirectional encoder representation for transformer

来源：评论

学校读者我要写书评

暂无评论

Learning Decision Policies with Instrumental Variables through Double Machine Learning 41

Learning Decision Policies with Instrumental Variables throu...

引用

41st International Conference on Machine Learning, ICML 2024

作者： Shao, Daqian Soleymani, Ashkan Quinzan, Francesco Kwiatkowska, Marta Department of Computer Science University of Oxford United Kingdom Department of Electrical Engineering and Computer Science Massachusetts Institute of Technology United States

A common issue in learning decision-making policies in data-rich settings is spurious correlations in the offline dataset, which can be caused by hidden confounders. Instrumental variable (IV) regression, which utilises a key unconfounded variable known as the instrument, is a standard technique for learning causal relationships between confounded action, outcome, and context variables. Most recent IV regression algorithms use a two-stage approach, where a deep neural network (DNN) estimator learnt in the first stage is directly plugged into the second stage, in which another DNN is used to estimate the causal effect. Naively plugging the estimator can cause heavy bias in the second stage, especially when regularisation bias is present in the first stage estimator. We propose DML-IV, a non-linear IV regression method that reduces the bias in two-stage IV regressions and effectively learns high-performing policies. We derive a novel learning objective to reduce bias and design the DML-IV algorithm following the double/debiased machine learning (DML) framework. The learnt DML-IV estimator has strong convergence rate and O(N−1/2) suboptimality guarantees that match those when the dataset is unconfounded. DML-IV outperforms state-of-the-art IV regression methods on IV regression benchmarks and learns high-performing policies in the presence of instruments. Copyright 2024 by the author(s)

关键词： Contrastive Learning

来源：评论

学校读者我要写书评

暂无评论

Design of an Optical Wireless Communication Link Using Mid Wave Infrared Laser Source & Comparison with Short Wave Infrared Laser Sources in Different Atmospheric Anomalies: An Analytical Study

引用

Nonlinear Optics Quantum Optics 2024年第1-2期60卷 85-111页

作者： Mukherjee, Shibabrata Paul, Sujoy Mazumdar, Saswati Electrical Engineering Department Jadavpur University West Bengal Kolkata700032 India Centre for Distributed Computing Computer Science Engineering Department Jadavpur University West Bengal Kolkata700032 India

Optical Wireless Communication (OWC), sometimes referred to as Free Space Optics (FSO), is the backbone of leading-edge communication systems that offer high performance and fast speed. The main pitfall of this system is that the optical signal is hindered by the different atmospheric anomalies. As a result, the system performance has been distorted and degraded. Among the different atmospheric anomalies like fog, rain, snow, etc., fog is the main enemy of this system. Using the properly chosen optics at the transmitter and receiver sides and wavelength, the system performance may be increased in diverse weather conditions. In this article, four wavelengths like two Mid Wave Infrared (MWIR) (3400 nm, 4100 nm) and two Short Wave Infrared (SWIR) (1550 nm & 2000 nm), 27 dBm optical peak power, the OOK modulation technique, 10 MHz modulation bandwidth, 10-7 BER, 2 km optical wireless link distance have been considered. From the theoretical simulation results, it has been initiated that MWIR optical sources (like 4100 nm and 3400 nm) contribute 72.22 dB, and 71.27 dB link margin at adverse atmospheric anomalies respectively, whereas SWIR Optical sources like 1550 nm & 2000 nm furnish 67.70 dB, 68.50 dB link margin respectively for 2 km link distance. The 4100 nm wavelength Laser source performs better in different atmospheric anomalies and provides a better pointing error angle in clear, as well as adverse weather conditions, compared to other wavelengths. © 2024 Old City Publishing, Inc.

关键词： Space optics

来源：评论

学校读者我要写书评

暂无评论

Highly Efficient Three-Level AC-AC Converter with Identical In-Phase and Antiphase Buck-Boost Operations

引用

IEEE Transactions on Power Electronics 2024年第8期39卷 9929-9942页

作者： Ahmed, Hafiz Furqan Khan, Ashraf Ali Al Zaabi, Omar Mousavi, Seyyed Mohammad Javad Babaei, Ebrahim National Sun Yat-Sen University Department of Electrical Engineering Kaohsiung804 Taiwan Memorial University of Newfoundland Department of Electrical and Computer Engineering Faculty of Engineering and Applied Science St. John'sNLA1C 5S7 Canada Khalifa University Advanced Power and Energy Center Department of Electrical Engineering and Computer Science Abu Dhabi127788 United Arab Emirates University of Tabriz Faculty of Electrical and Computer Engineering Tabriz51666 Iran Near East University Engineering Faculty Nicosia99138 Turkey

In this article, a new coupled-inductors based three-level bipolar buck-boost ac-ac converter is proposed. The proposed converter can produce highly efficient and symmetric in-phase and antiphase buck and boost modes of operation with much lower component voltage and current stresses. This reduces the operation and design complexity, and helps to achieve a high-efficiency operation. The high-frequency modulating switches of the proposed converter are implemented with coupled-inductor-based dual-buck phase leg, producing three-level input and output voltages, and providing inherent protection from voltage-source short-circuit issue during switch transitions. The other salient features of the proposed converter are no-commutation issue, no need for RC snubbers or dedicated safe-commutation algorithms, provision of high-quality and continuous input/output currents, and support for nonresistive loads. An in-depth circuit analysis of the proposed converter is provided based on the proposed switch modulation strategies. The guidelines for component design/selection are discussed, followed by the comparisons with state-of-the-art three-level ac-ac converters. Finally, practical circuit verifications are performed on a laboratory-assembled prototype. © 1986-2012 IEEE.

关键词： Buck-Boost converter

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：