检索结果-内蒙古大学图书馆

IEEE International Conference on Smart Internet of Things (SmartIoT)

作者： Tingjing Nie Xi Wang Yong Zhang Department of Engineering Shenzhen MSU-BIT University Shenzhen China Artificial Intelligence Research Institute Shenzhen MSU-BIT University Shenzhen China

ISBN: (数字)9798350366440

ISBN: (纸本)9798350366457

Due to the rapid development of artificial intelli-gence and Internet of Things technologies, the use of large-scale networks and large-scale data has made single-GPU training insufficient to meet demands. Therefore, distributed training of large models using multiple nodes is a necessary approach. At the same time, the high-frequency communication among multiple GPUs has slowed down the speed of distributed training. This study aims to increase the interaction cycles among distributed training nodes to reduce communication overhead, thereby accelerating distributed training. We will compare the classification accuracy and convergence speed of neural networks under different interaction cycles and stochastic optimization algorithms using the MNIST dataset and a single fully connected layer architecture, to inspire subsequent related work.

关键词： Training Accuracy Computational modeling Neural networks Stochastic processes Distributed databases classification algorithms Internet of Things Optimization Convergence

来源：评论

学校读者我要写书评

暂无评论

A Novel One-Versus-All Approach for Multiclass classification in TinyML Systems

引用

IEEE EMBEDDED SYSTEMS LETTERS 2025年第2期17卷 71-74页

作者： Puslecki, Tobiasz Walkowiak, Krzysztof Wroclaw Univ Sci & Technol Dept Syst & Comp Networks PL-50370 Wroclaw Poland

The recent progress in TinyML technologies triggers the need to address the challenge of balancing inference time and recognition quality. TinyML systems are defined by specific constraints in computation, memory and energy. These constraints emphasize the need for specialized optimization techniques when implementing machine learning (ML) applications on such platforms. While deep neural networks are popular in TinyML systems, exploring simple classifiers is also worthwhile. In this work, we consider a modification of the one-versus-all (OVA) approach in a multiclass task of computer vision in TinyML systems. This modification, named thresholded OVA (TOVA), enables control over classification accuracy, influencing both latency and energy consumption per inference. By testing various combinations of hyperparameters, we simulate the performance of a real device using metrics specific to TinyML systems. The results show that the proposed method significantly saves energy and speeds up computation, at the cost of slightly lower-overall accuracy of the TinyML system.

关键词： Tiny machine learning Accuracy Computational modeling Fuzzy sets Batteries Energy consumption Performance evaluation Indexes classification algorithms Training Binary classifiers embedded systems energy consumption fuzzy logic TinyML

来源：评论

学校读者我要写书评

暂无评论

Patient Health classification Using Various Machine Learning algorithms

Patient Health Classification Using Various Machine Learning...

引用

Control Instrumentation System Conference (CISCON)

作者： Aryan Rahul Shah Ammu Anna Mathew Karthikeyan V Vivekanandan Shanmugasundaram School of Electrical Engineering Vellore Institute of Technology Vellore India

ISBN: (数字)9798350375480

ISBN: (纸本)9798350375497

This paper analyzes the Body Performance dataset using several machine learning algorithms to evaluate individuals’ health according to their 11 physical characteristics. The dataset contains data on age, weight, height, body fat percentage, blood pressure, and other factors. The machine learning techniques used are Linear regression, gradient boosting, Gaussian Naive Bayes, Random Forest, and Support Vector Machines. Each algorithm's performance is assessed using its accuracy, precision, F1 score, and confusion matrix. We observed that both the Random Forest and the Gaussian Naive Bayes provide an accuracy of 85.714%, with the Gaussian Naive Bayes performing the computation in 0.011 seconds as opposed to 69.253 seconds for the Random Forest. We also observed that Linear Regression gives an accuracy of 71.428%, whereas both Gradient Boosting and SVM have an accuracy of 78.571%. We have concluded that the Random Forest Algorithm is a better choice because of both a higher F1 score and a higher Precision value. These algorithms could be used to construct a diagnostic tool for medical practitioners to assist with patient classification.

关键词： Support vector machines Machine learning algorithms Accuracy Linear regression Boosting Vectors classification algorithms Bayes methods Random forests Medical diagnostic imaging

来源：评论

学校读者我要写书评

暂无评论

Evaluating Deep Learning algorithms for MRI-Based Brain Tumor classification

Evaluating Deep Learning Algorithms for MRI-Based Brain Tumo...

引用

Emerging Innovations and Advanced Computing (INNOCOMP), International Conference on

作者： Gurpreet Singh Amit Chhabra Ajay Mittal Department of Computer Science and Engineering Chandigarh College of Engineering and Technology (Degree Wing) Chandigarh India Department of Computer Science and Engineering University Institute of Engineering and Technology Panjab University Chandigarh India

ISBN: (数字)9798350376470

ISBN: (纸本)9798350376487

The classification of brain tumors has significant importance in the realm of clinical diagnosis and the implementation of appropriate treatment strategies. The process of diagnosing a brain tumor is time-consuming and strongly dependent on the expertise and experience of the radiologist. The volume of data that necessitates processing has seen a significant surge in parallel with the growth in patient population, rendering previous protocols financially burdensome and inefficient. Numerous researchers have conducted investigations on a range of algorithms aimed at the detection and classification of brain tumors, with a focus on achieving high levels of accuracy and computational efficiency. In recent times, there has been a surge in the utilization of Deep Learning (DL) methodologies to construct automated systems that possess the ability to effectively diagnose or segment brain tumors within a reduced timeframe. Deep learning (DL) facilitates the use of a pre-existing Convolutional Neural Network (CNN) model designed for the analysis of medical pictures, with a special focus on the classification of brain tumors. The rapid advancement of Deep Learning techniques in the context of computer vision tasks has been made possible by the availability of extensive training datasets and advancements in model architecture, leading to enhanced approximations within a supervised learning framework. The growth has mostly been motivated by the need to implement these ideas. The use of deep learning methods has shown potential in enhancing the accuracy of brain tumor identification and classification via the utilization of magnetic resonance imaging (MRI). According to the findings, CNN is a superior technique that greatly boosts the accuracy by 93.4% and it is maximum as compared to other methods.

关键词： Deep learning Training Accuracy Magnetic resonance imaging Computational modeling Brain modeling classification algorithms

来源：评论

学校读者我要写书评

暂无评论

Electrocardiogram classification Based on Spatial-Channel Networks and Optimization algorithms

Electrocardiogram Classification Based on Spatial-Channel Ne...

引用

Chinese Automation Congress (CAC)

作者： Xuewen Zhou Jiejie Chen Ping Jiang Xinrui Zhang School of Computer and Information Engineering Hubei Normal University Huangshi China School of Computer Hubei PolyTechnic University Huangshi China

ISBN: (数字)9798350368604

ISBN: (纸本)9798350368611

Cardiovascular diseases are one of the leading causes of death globally, and the electrocardiogram (ECG) serves as a primary diagnostic tool, making automated ECG classification of significant clinical importance. Although previous studies have extensively utilized various neural network models for feature extraction and classification of ECG signals, these methods often face challenges such as insufficient accuracy, high computational demands, and lengthy processing times. This study aims to design a new arrhythmia classification framework that achieves high accuracy and low cost in diagnosing arrhythmias. We introduced Spatial and Channel-wise Convolutional (SCC) attention into Temporal Convolutional Networks (TCN), resulting in the SCC-TCN model, which significantly enhances classification accuracy and reduces computational costs. Additionally, by incorporating a stochastic walking mechanism, we improved the Rime optimization algorithm(RIME), effectively enhancing its performance. We also utilized the improved RIME algorithm to optimize the hyperparameters of the Spatial and Channel-wise Temporal Convolutional Networks(SCC-TCN) model, further improving its classification performance. Testing on the MIT-BIH Arrhythmia Dataset (MITAD) and the MIT-BIH Supraven-tricular Arrhythmia Dataset (SVDB) demonstrated that our method achieves classification accuracies of 99.40 % and 97.82%, respectively, validating the practical value of this approach.

关键词： Accuracy Convolution Arrhythmia Computational modeling Stochastic processes Electrocardiography Feature extraction classification algorithms Optimization Testing

来源：评论

学校读者我要写书评

暂无评论

Enhanced Machine Learning algorithms for classification in Agricultural Advisory Systems and Agricultural Data Analysis

Enhanced Machine Learning Algorithms for Classification in A...

引用

Electronic Systems and Intelligent Computing (ICESIC), International Conference on

作者： Logesh M Domnic S J. Nelson Raja Electrical & Electronics Engineering (SG & EV) Sastra Deemed University Thanjavur Tamil Nadu India Department of Computer Application National Institute of Technonology Tiruchirappalli Tamil Nadu India Department of Computer Science Thiagarajar College Madurai Tamil Nadu India

ISBN: (数字)9798331522988

ISBN: (纸本)9798331522995

Agricultural technology, ensuring reliability, and optimizing processes are paramount. This work explores the application of machine learning algorithms (ML) to two critical areas: agricultural advisory systems (AAS) and agricultural data analysis (ADA). By leveraging historical data and advanced algorithms, this work aims to enhance predictive accuracy and operational efficiency in ADA and AAS. The proposed work focuses on various ML models for classification tasks, comparing their performance metrics, and discussing their applicability in agricultural sector. The findings highlight the ML model to significantly improve outcomes in agricultural management.

关键词： Measurement Machine learning algorithms Data analysis Accuracy Computational modeling Prediction algorithms classification algorithms Reliability

来源：评论

学校读者我要写书评

暂无评论

Aircraft-Type classification Using Deep Learning algorithms

Aircraft-Type Classification Using Deep Learning Algorithms

引用

International IEEE Conference on Intelligent Systems, IS

作者： Noora Alkharji Hessa Almazrouei Shaikha Alzaabi Ali Bou Nassif Mariame Elsalhy Manar Abu Talib Department of Computer Engineering University of Sharjah Sharjah UAE Department of Computer Science University of Sharjah Sharjah UAE

ISBN: (数字)9798350350982

ISBN: (纸本)9798350350999

In the field of artificial intelligence (AI), deep learning (DL) has become an essential methodology with many uses, particularly in the aviation sector. The efficiency and effectiveness of tasks connected to aircraft recognition and classification are considerably improved by DL. The goal of this study is to examine and deal with problems related to efficiency and accuracy in DL models for identifying and classifying aircrafts. In this article, we suggest a solution that utilizes a customized convolutional neural network (CNN) model that our team has created and assess its performance in comparison to other deep learning (DL) models like VGG16 and ResNet50. Accuracy, recall, precision, and f1-score are among the metrics used for evaluation. The outcomes of this study demonstrate significant gains and offer insightful information for future endeavors in the aviation field.

关键词： Deep learning Accuracy Atmospheric modeling Military aircraft Data models classification algorithms Convolutional neural networks Noise measurement Aircraft Residual neural networks

来源：评论

学校读者我要写书评

暂无评论

Optimized Feature Selection and classification Framework for Cardiovascular Disease Using Statistical Normalization and Bio-Inspired algorithms

Optimized Feature Selection and Classification Framework for...

引用

Communication, Control and Intelligent Systems (CCIS)

作者： N. Muthumani K. Akilandeswari Department of Computer Science PPG College of Arts and Science Coimbatore Tamil Nadu

ISBN: (数字)9798331528201

ISBN: (纸本)9798331528218

Heart disease, which includes various circulatory illnesses that largely impact the structure and function of the heart, is a prevalent primary cause of morbidity and mortality. This long-term illness significantly lowers people’s quality of life and raises healthcare costs. This illness will cause the body to react severely, including heart attacks and cardiac arrest, which may outcome in excruciating losses or even death. Early diagnosis is crucial for survival and treatment effectiveness in order to prevent severe consequences. Some medical diagnostic techniques, such as ECG, ECO, and stress tests, are available for the identification of cardiovascular illness; however, these techniques have certain drawbacks and restrictions caused by challenges with cost, personnel, and accessibility. Remedies are thus required to get over these drawbacks, which are satisfied by computer algorithms. Machine learning algorithms meet this requirement by combining various data types, handling metadata, and identifying medical risk factors and patterns that traditional methods are unable to identify. The Ant-Hybrid Support Vector Machine Learning approach, which employs the hybrid kernel support vector machine learning algorithm for disease classification and Ant colony optimization for feature extraction, is used in this research work to enhance the capabilities of feature selection and classification. This technique will improve feature extraction and classification capabilities by identifying intricate patterns in the dataset that are overlooked by current approaches. The performance of the algorithms for the prediction of cardiovascular disease is compared using the different performance metrics such as Accuracy, Precision, Recall and this comparison clearly shows that the proposed model outperforms the existing algorithms with the highest accuracy of 98%. To increase model accuracy, the z-Score Normalization technique is employed during the preprocessing step to normalize the

关键词： Support vector machines Heart Measurement Machine learning algorithms Accuracy Biological system modeling Feature extraction Prediction algorithms classification algorithms Optimization

来源：评论

学校读者我要写书评

暂无评论

Study of the Application of Logistic Regression and Naïve Bayes algorithms for Automatic classification of User Reviews in Bulgarian

Study of the Application of Logistic Regression and Naïve B...

引用

International Symposium on Multidisciplinary Studies and Innovative Technologies (ISMSIT)

作者： Irena Valova Gabriel Kanev Tsvetelina Kaneva Department of Computer Systems and Technologies University of Ruse Ruse Bulgaria

ISBN: (数字)9798350354423

ISBN: (纸本)9798350354430

Sentiment analysis (SA) is a critical task in natural language processing (NLP), aimed at evaluating and classifying user opinions in the form of text. This paper explores various methods for constructing sentiment lexicons, including manual, corpus-based, and dictionary-based approaches, each with its strengths and limitations. A specific focus is given to the challenges and methods of analyzing Bulgarian texts, highlighting the linguistic complexities that impact sentiment analysis in this language. The study includes an experiment using Logistic Regression (LR) and Naïve Bayes (NB) classifiers on a Bulgarian dataset, with vectorization techniques such as CountVectorizer and TF-IDF. The results are compared and examined and the combination of an algorithm and vectorization most suitable for this Bulgarian sentiment analysis is determined.

关键词： Analytical models Sentiment analysis Logistic regression Accuracy Reviews Manuals Machine learning Bayes methods Complexity theory classification algorithms

来源：评论

学校读者我要写书评

暂无评论

Software Fault Detection algorithms Using Artificial Intelligence: A Review and classification

Software Fault Detection Algorithms Using Artificial Intelli...

引用

Electrical, Computer and Energy Technologies (ICECET), International Conference on

作者： Lungisani Ndlovu Bheki Cossa Caroline Makokoe Lethukuthula Khumalo Ntomfuthi Ntshangase Siphelele Myaka Nomalisa Ndhlovu Mamello Mtshali Information and Cyber Security Centre Council for Scientific and Industrial Research (CSIR) Pretoria South Africa

ISBN: (数字)9798350395914

ISBN: (纸本)9798350395921

With the rise of software applications, software development has become a rapidly evolving field, propelled by technological advancements and the increasing demand for innovative solutions to sustain our digital era. Detecting software faults has emerged as a critical attribute in the software development cycle to ensure system reliability, quality, and user satisfaction. However, traditional fault detection methods often suffer from drawbacks such as time consumption and error proneness, particularly in the context of large and complex software systems. In response to these hurdles, artificial intelligence (AI) presents itself as a promising methodology, with the aim of improving accuracy, scalability, automation, and proactiveness in fault detection. This paper systematically reviews the literature on AI-based software fault detection algorithms. Leveraging the PRISMA framework (Preferred Reporting Items for Systematic Reviews and Meta-Analyses), a curated set of 50 peer-reviewed research publications was identified in Google Scholar and subsequently analyzed using a classification coding framework. The findings reveal that AI-based software fault detection algorithms played an important role in improving software system reliability and performance.

关键词： Fault detection Scalability Software algorithms Propulsion Software systems classification algorithms Software reliability Internet Artificial intelligence Software development management

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：