检索结果-内蒙古大学图书馆

Fast Learning Network Algorithm for Voice Pathology Detection and Classification

Multimedia Tools and Applications 2025年第17期84卷 18567-18598页

作者： Albadr, Musatafa Abbas Abbood Ayob, Masri Tiun, Sabrina AL-Dhief, Fahad Taha Al-Daweri, Muataz Salam Homod, Raad Z. Abbas, Ali Hashim Department of Petroleum Project Management College of Industrial Management of Oil and Gas Basrah University for Oil and Gas Al-Basrah Iraq Faculty of Information Science and Technology CAIT Universiti Kebangsaan Malaysia Selangor Bangi Malaysia Department of Communication Engineering School of Electrical Engineering Universiti Teknologi Malaysia UTM Johor Bahru Johor Malaysia Department of Information Systems College of Economics Management & Information Systems University of Nizwa Nizwa Oman Department of Oil and Gas Engineering Basrah University for Oil and Gas Al-Basrah Iraq Department of Computer Technical Engineering College of Information Technology Imam Ja’afar Al-Sadiq University Al-Muthanna Iraq

The utilisation of ML (Machine Learning) techniques in the detection of the VP (Voice Pathology) has recently gained a lot of consideration. However, these efforts still have several drawbacks such as: I) the accuracy rates of most previous works are still not promising and need more improvement. II) The majority of the previous works have concentrated on the task of VP detection only and disregarded the task of VP classification. III) Most of the previous works have been assessed based on a single voice data only like the vowel /a/, and the other vowels (/i/, and /u/) and sentences were disregarded. IV) The majority of the previous works performance were assessed utilising a limited evaluation metrics. Recently, one of the utmost effective ML techniques is FLN (Fast Learning Network), it is an efficient technique for data classification. However, the FLN classifier has not been implemented to the problem of VP detection and classification. Therefore, this research proposes the FLN classifier with MFCC (Mel-Frequency Cepstral Coefficient) features in order to enhance the accuracy of the VP detection and classification. The FLN classifier has the ability to a) solve both binary and multiclass classification issues, b) eliminate overfitting, as well as c) operate akin to a NN (Neural Network) structure while employing the principles of a kernel-based SVM (Support Vector Machine). In this research, the SVD (Saarbrucken Voice Database) dataset was utilised to assess the FLN classifier performance in the VP detection and classification. The assessment of the proposed FLN classifier was conducted based on two phases: the first phase includes all the voice samples of the SVD dataset with the sentences and vowels (i.e., /a/, /i/, and /u/) which are pronounced in neutral, high, and low pitches. While the second phase uses the voice samples of the utmost common 3 pathology types (i.e., paralysis, polyp, and cyst) based on the vowel /a/ that is pronounced in neutral pitch. The

关键词： Classification (of information)

来源：评论

学校读者我要写书评

暂无评论

Using Interim Recommitment to Reduce the Operational-cost Impacts of Wind Uncertainty

引用

Journal of Modern Power systems and Clean Energy 2022年第4期10卷 839-849页

作者： Mahan A.Mansouri Ramteen Sioshansi Department of Integrated Systems Engineering The Ohio State UniversityColumbusOH 43210USA Department of Integrated Systems Engineering Department of Electrical and Computer Engineeringand Center for Automotive ResearchThe Ohio State UniversityColumbusOH 43210USA

Using wind-availability forecasts in day-ahead unit commitment can require expensive real-time operational *** examine the benefit of conducting interim recommitment between day-ahead unit commitment and real-time *** a simple stylized example and a case study that is based on ISO New England,we compare system-operation costs with and without interim *** find an important tradeoff—later recommitment provides better wind-availability forecasts,but the system has less flexibility due to operating *** the time windows that we examine,hour-20 recommitment provides the greatest operational-cost reduction.

关键词： Power-system operation power-system economics unit commitment economic dispatch wind generation.

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

A lightweight deep learning model with knowledge distillation for pulmonary diseases detection in chest X-rays

引用

Multimedia Tools and Applications 2025年第15期84卷 14885-14913页

作者： Asham, Mohammed A. Al-Shargabi, Asma A. Al-Sabri, Raeed Meftah, Ibrahim School of Computer Science and Engineering Central South University Hunan Changsha410083 China Department of Information Technology Collage of Computer Qassim University Buraydah51452 Saudi Arabia College of Mechanical and Electrical Engineering Central South University Changsha410017 China

Accurate and timely diagnosis of pulmonary diseases is critical in the field of medical imaging. While deep learning models have shown promise in this regard, the current methods for developing such models often require extensive computing resources and complex procedures, rendering them impractical. This study focuses on the development of a lightweight deep-learning model for the detection of pulmonary diseases. Leveraging the benefits of knowledge distillation (KD) and the integration of the ConvMixer block, we propose a novel lightweight student model based on the MobileNet architecture. The methodology begins with training multiple teacher model candidates to identify the most suitable teacher model. Subsequently, KD is employed, utilizing the insights of this robust teacher model to enhance the performance of the student model. The objective is to reduce the student model's parameter size and computational complexity while preserving its diagnostic accuracy. We perform an in-depth analysis of our proposed model's performance compared to various well-established pre-trained student models, including MobileNetV2, ResNet50, InceptionV3, Xception, and NasNetMobile. Through extensive experimentation and evaluation across diverse datasets, including chest X-rays of different pulmonary diseases such as pneumonia, COVID-19, tuberculosis, and pneumothorax, we demonstrate the robustness and effectiveness of our proposed model in diagnosing various chest infections. Our model showcases superior performance, achieving an impressive classification accuracy of 97.92%. We emphasize the significant reduction in model complexity, with 0.63 million parameters, allowing for efficient inference and rapid prediction times, rendering it ideal for resource-constrained environments. Outperforming various pre-trained student models in terms of overall performance and computation cost, our findings underscore the effectiveness of the proposed KD strategy and the integration of the Conv

关键词： Diagnosis

来源：评论

学校读者我要写书评

暂无评论

Sigmoidal Particle Swarm Optimization for Twitter Sentiment Analysis

引用

computers, Materials & Continua 2023年第1期74卷 897-914页

作者： Sandeep Kumar Muhammad Badruddin Khan Mozaherul Hoque Abul Hasanat Abdul Khader Jilani Saudagar Abdullah AlTameem Mohammed AlKhathami Department of Computer Science and Engineering CHRIST(Deemed to be University)Bangalore560074India Information Systems Department College of Computer and Information SciencesImam Mohammad Ibn Saud Islamic University(IMSIU)Riyadh11432Saudi Arabia

Social media,like Twitter,is a data repository,and people exchange views on global issues like the COVID-19 *** media has been shown to influence the low acceptance of *** work aims to identify public sentiments concerning the COVID-19 vaccines and better understand the individual’s sensitivities and feelings that lead to *** work proposes a method to analyze the opinion of an individual’s tweet about the COVID-19 *** paper introduces a sigmoidal particle swarm optimization(SPSO)***,the performance of SPSO is measured on a set of 12 benchmark problems,and later it is deployed for selecting optimal text features and categorizing *** proposed method uses TextBlob and VADER for sentiment analysis,CountVectorizer,and term frequency-inverse document frequency(TF-IDF)vectorizer for feature extraction,followed by SPSO-based feature *** Covid-19 vaccination tweets dataset was created and used for training,validating,and *** proposed approach outperformed considered algorithms in terms of ***,we augmented the newly created dataset to make it balanced to increase performance.A classical support vector machine(SVM)gives better accuracy for the augmented dataset without a feature selection *** shows that augmentation improves the overall accuracy of tweet *** the augmentation performance of PSO and SPSO is improved by almost 7%and 5%,respectively,it is observed that simple SVMwith 10-fold cross-validation significantly improved compared to the primary dataset.

关键词： Twitter data analysis sentiment analysis social media analytics swarm intelligence COVID-19 vaccine

来源：评论

学校读者我要写书评

暂无评论

Enhancement of UAV Data Security and Privacy via Ethereum Blockchain Technology

引用

computers, Materials & Continua 2023年第8期76卷 1797-1815页

作者： Sur Singh Rawat Youseef Alotaibi Nitima Malsa Vimal Gupta Department of Computer Science and Engineering JSS Academy of Technical EducationNoida201301India Department of Computer Science College of Computer and Information SystemsUmm Al-Qura UniversityMakkah21955Saudi Arabia

Unmanned aerial vehicles(UAVs),or drones,have revolutionized a wide range of industries,including monitoring,agriculture,surveillance,and supply ***,their widespread use also poses significant challenges,such as public safety,privacy,and ***,targetingUAVs have become more frequent,which highlights the need for robust security *** technology,the foundation of cryptocurrencies has the potential to address these *** study suggests a platform that utilizes blockchain technology tomanage drone operations securely and *** incorporating blockchain technology,the proposed method aims to increase the security and privacy of drone *** suggested platform stores information on a public blockchain located on Ethereum and leverages the Ganache platform to ensure secure and private blockchain *** wallet for Ethbalance is necessary for BCT *** present research finding shows that the proposed approach’s efficiency and security features are superior to existing *** study contributes to the development of a secure and efficient system for managing drone operations that could have significant applications in various *** proposed platform’s security measures could mitigate privacy concerns,minimize cyber security risk,and enhance public safety,ultimately promoting the widespread adoption of *** results of the study demonstrate that the blockchain can ensure the fulfillment of core security needs such as authentication,privacy preservation,confidentiality,integrity,and access control.

关键词： Unmanned aerial vehicles(UAVs) blockchain data privacy network security smart contract Ethereum

来源：评论

学校读者我要写书评

暂无评论

Combination of Interleaved Single-input Multiple-output DC-DC Converters

引用

CSEE Journal of Power and Energy systems 2022年第1期8卷 132-142页

作者： Eladio Dur´an Aranda Salvador P´erez Litr´an Mar´ıa Bella Ferrera Prieto Electronic Engineering Computer Systems and Automation DepartmentUniversity of HuelvaHuelvaSpain Electrical and Thermal Engineering Design and Projects DepartmentUniversity of HuelvaHuelvaSpain

This paper analyses,simulates and verifies an experimental prototype of a four-phase interleaved DC-DC *** is based on a SEPIC-Cuk *** developed prototype has been used in single-input multiple-output(SIMO)*** combined converter allows obtaining dual output voltages of the same value,from a single input DC voltage and with only a power *** interleaved DC-DC converters achieve a better dynamic response and low ripple,maintaining their *** converter is connected in parallel,thereby managing their losses by distributing them between more components,which facilitates the thermal management of the multiphase converter and allows handling high power values in small sizes with respect to solutions for a single *** control strategies were applied:synchronous operation mode(SOM)and interleaved operation mode(IOM).The simulation results allow the comparison of both operational modes,verifying that the IOM presents advantages with respect to the ripple at the input and output *** experimental prototype was designed for a distributed power architecture and bipolar DC microgrid(MG).

关键词： Bipolar DC DC-DC converter interleaved

来源：评论

学校读者我要写书评

暂无评论

Joint Optimizations for Double-IRS' Cooperative Positioning and Beamforming Over Massive-MIMOAP Based 6G Secure Mobile Wireless Networks

Joint Optimizations for Double-IRS' Cooperative Positioning ...

引用

2024 IEEE Global Communications Conference, GLOBECOM 2024

作者： Wang, Jiaojie Wang, Fei Zhang, Xi Yang, Yuanyuan Southwest University College of Electronic and Information Engineering Chongqing400715 China Texas A&m University Networking and Information Systems Laboratory Department of Electrical and Computer Engineering College StationTX77843 United States Stony Brook University Department of Electrical and Computer Engineering Stony BrookNY11794 United States

ISBN: (纸本)9798350351255

Intelligent reflecting surface (IRS) has been widely recognized as one of the key techniques to improve secure communications performances. However, most existing works mainly focus on the passive beamforming, i.e., phase-shifts, design of a single IRS, without taking into account the cooperations among multiple IRSs and the optimizations of their relative positions. To overcome these deficiencies, in this paper we propose the joint optimizations between the transmit beamforming of massive multiple-input multiple-output (massive-MIMO) access point (AP) and double-IRS' positions and passive beamforming over 6G secure mobile networks. In our proposed networking architectures, AP transmits data to multiple mobile users (MUs) through reflections of two cooperative IRSs under the existence of one eavesdropper (Eve). First, we formulate a secrecy rate optimization problem to maximize the expectations of all MUs' aggregate secrecy rates when Eve's exact position is unknown. Second, leveraging the deep reinforcement learning (DRL), we develop two joint deploying and beamforming schemes to tackle the uncertainty of Eve's exact position. Finally, we validate and evaluate our developed schemes by conducting the extensive simulations, showing the significant performances improvements of our developed schemes through jointly optimizing IRSs' positions/orientations and transmit-beamforming of massive-MIMOAP as the function of the predicted Eve's deploying areas. © 2024 IEEE.

关键词： Deep reinforcement learning

来源：评论

学校读者我要写书评

暂无评论

Machine Learning for Smart Agriculture: A Comprehensive Survey

IEEE Transactions on Artificial Intelligence

引用

IEEE Transactions on Artificial Intelligence 2024年第6期5卷 2568-2588页

作者： Mahmood, M. Rezwanul Matin, Mohammad Abdul Goudos, Sotirios K. Karagiannidis, George North South University Department of Electrical and Computer Engineering Dhaka1229 Bangladesh Aristotle University of Thessaloniki ELEDIA@AUTH Department of Physics Thessaloniki54124 Greece Aristotle University of Thessaloniki Department of Electrical and Computer Engineering Thessaloniki54124 Greece Cyber Security Systems and Applied AI Research Center Beirut1102 2801 Lebanon

As communication technologies and equipment evolve, smart assets become smarter. The agricultural industry is also evolving in line with the implementation of modern communication protocols, intelligent sensors, and equipment. This evolution is enabling large-scale agricultural production processes to operate independently, thus, securing the food supply chain for an ever-growing population. Data processing for such a system with multiple heterogeneous sources requires proper management for effective agricultural operations. Recognizing the advantages of machine learning (ML) in performing large-scale data processing, researchers are investigating the implementation of ML to design an effective intelligent agricultural architecture. The aim of this article is to provide a thorough analysis of the state-of-the-art in smart agriculture, open challenges, and guidelines for the development of further enhanced smart agriculture systems. Specifically, we describe how ML is used to create intelligent agricultural systems supported by state-of-the-art technology. © 2020 IEEE.

关键词： Weed control

来源：评论

学校读者我要写书评

暂无评论

A Proactive Crowdsourcing Framework for Fake Base Station Detection and Avoidance 15

A Proactive Crowdsourcing Framework for Fake Base Station De...

引用

15th IEEE Annual Computing and Communication Workshop and Conference, CCWC 2025

作者： Saad, Omar A. Etman, Ahmed A. Abdel-Malek, Mai A. Azab, Mohamed Alexandria University Faculty of Engineering Department of Computer and Communications Engineering Alexandria Egypt The University of Arizona Department of Electrical and Computer Engineering AZ United States Virginia Military Institute Department of Computer and Information Sciences VA United States

ISBN: (纸本)9798331507695

With next-generation communication technologies becoming more integrated into critical infrastructure, securing these networks against sophisticated threats has become paramount. One of the most persistent threats to this communication is the growing presence of fake base stations (FBS). These FBSs are malicious entities that pose as legitimate base stations to intercept, manipulate, or exploit mobile communications, leading to serious privacy breaches. While some existing literature addresses the detection of such malicious entities, detecting FBSs has traditionally been the responsibility of network operators, who rely on tools like signal strength analysis and radio frequency (RF) monitoring to identify unusual patterns or rogue signals. However, these centralized methods often struggle to detect FBSs in complex environments characterized by high mobility, dense urban areas, and sophisticated spoofing techniques. This paper proposes a scalable and pervasive blockchain-based crowd-sourcing mechanism to enable extensive device participation in identifying FBSs. In a fully automated and distributed manner, the proposed platform uses vetted and verified information recorded on the blockchain to identify and report suspicious FBSs in the area. This allows other users to dissociate from these malicious entities automatically. A reputation management system is also employed to ensure the reliability and accountability of participating users. Using qualitative and quantitative evaluations, the proposed approach demonstrated effectiveness and efficiency in achieving all targeted goals. © 2025 IEEE.

关键词： Crowdsourcing

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：