检索结果-内蒙古大学图书馆

International Conference on Computing and Networking Technology (ICCNT)

作者： Ayain John Navya Damodar N Mohith Bharathy Vijayan Dept. Artificial Intelligence and Machine Learning Dayananda Sagar University Bangalore India Dept. Computer Science Engineering Dayananda Sagar University Bangalore India

ISBN: (数字)9798350370249

ISBN: (纸本)9798350370270

The research work investigates the factors that influence the quality of red wine and how they contribute to positive outcomes. It uses machine learning techniques like Support Vector machines (SVM) and K-nearest neighbors (KNN) to classify and assess red wine quality data. Traditionally, red wine quality has been evaluated subjectively, which presents challenges. machine learning algorithms offer an objective alternative. This study compares SVM and KNN algorithms in assessing red wine quality and aims to determine the crucial attributes that affect its quality. To evaluate the classification performance of different machine learning algorithms, metrics like specificity, sensitivity, selectivity, and F1 measure are used. Additionally, grid search algorithms are employed for parameter optimization. In the red wine datasets, the Support Vector machine (SVM) model obtains an accuracy rate of 89%, surpassing the K-Nearest Neighbors (KNN) model’s accuracy rate of 87%, especially regarding predictive specificity. Furthermore, using grid search algorithms improves the overall performance of the classifiers. This study highlights the effectiveness of machine learning in assessing red wine quality, providing valuable information for producers to optimize wine production processes and ensure high-quality products for consumers.

关键词： Support vector machines machine learning algorithms Accuracy Sensitivity Production Nearest neighbor methods Predictive models Prediction algorithms Vectors Classification algorithms

来源：评论

学校读者我要写书评

暂无评论

AI-Driven Diagnostics: Comparative Analysis of machine learning and Deep learning Approaches in Multiple Sclerosis Diagnostics

AI-Driven Diagnostics: Comparative Analysis of Machine Learn...

引用

International Conference on Advanced computer Control, ICACC

作者： Anitha L. D. Narmadha G. Naveen Sundar Division of Computer Science and Engineering Karunya Institute of Technology and Sciences Coimbatore India Division of Artificial Intelligence and Machine Learning Karunya Institute of Technology and Sciences Coimbatore India

ISBN: (数字)9798350379136

ISBN: (纸本)9798350379143

Multiple Sclerosis (MS) is a chronic autoimmune disorder that targets the central nervous system, resulting in considerable physical and cognitive disabilities. Timely and precise diagnosis is essential for the effective management and treatment of Multiple Sclerosis (MS). Predicting MS involves using various types of data to assess the risk and diagnose the disease. The data commonly used are clinical, imaging, speech and gene data. This work presents a review of machine learning and deep learning methods, encompassing a broad spectrum of algorithms such as Support Vector machines (SVM), Random Forest (RF), eXtreme Gradient Boosting (XGBoost), Convolutional Neural Networks (CNNs), and artificial Neural Networks (ANNs). Furthermore, the survey gives a comparative analysis of the dataset used, algorithms employed and the performance metrics. The survey also compares the implemented results with DL using the gene data. It also identifies current challenges in MS prediction, such as heterogeneity of MS, integration of multimodal data, validation and clinical implementation. Our goal is to offer a valuable resource for researchers and practitioners.

关键词： Deep learning Surveys Support vector machines Measurement Multiple sclerosis machine learning algorithms Reviews Prediction algorithms Reliability Random forests

来源：评论

学校读者我要写书评

暂无评论

learning Robust Global-Local Representation From EEG for Neural Epilepsy Detection

IEEE Transactions on Artificial Intelligence

引用

IEEE Transactions on artificial intelligence 2024年第11期5卷 5720-5732页

作者： Zhou, Xinliang Liu, Chenyu Yang, Ruizhi Zhang, Liangwei Zhai, Liming Jia, Ziyu Liu, Yang Nanyang Technological University School of Computer Science and Engineering 639798 Singapore Queen Mary University of London School of Electronic Engineering and Computer Science LondonE1 4NS United Kingdom Carnegie Mellon University School of Computer Science PittsburghPA15213 United States Central China Normal University Hubei Provincial Key Laboratory of Artificial Intelligence and Smart Learning Wuhan430079 China Central China Normal University School of Computer Science Wuhan430079 China Chinese Academy of Sciences Brainnetome Center Institute of Automation Beijing100190 China

Epilepsy is a life-threatening and challenging neurological disorder, and applying an electroencephalogram (EEG) is a commonly used clinical approach for its detection. Neuropsychological research indicates that epilepsy seizures are highly associated with distinct ranges of temporal EEG patterns. Although previous attempts to automatically detect epilepsy have achieved high classification performance, one crucial challenge still remains: how to effectively learn the robust global-local representation associated with epilepsy in the signals? To address the above challenge, we propose global-local neural epilepsy detection network (GlepNet), a novel architecture for automatic EEG epilepsy detection. We interleave the temporal convolution model together with the multihead attention mechanism within the GlepNet's encoder blocks to jointly capture the interlaced epilepsy seizure local and global features in EEG signals. Meanwhile, the interpretable method, gradient-weighted class activation mapping (Grad-CAM), is applied to visually confirm that the GlepNet acquires the ability to accord significant weight to EEG segments containing epileptiform abnormalities such as spike-wave complexes. Specifically, the Grad-CAM heatmaps are generated by backpropagating the gradients from the encoder blocks to highlight the epilepsy seizure-related parts. Extensive experiments show the superiority of the GlepNet over state-of-the-art methods on multiple EEG epilepsy datasets. The code will soon be open-sourced on GitHub. © 2024 IEEE.

关键词： Feature extraction

来源：评论

学校读者我要写书评

暂无评论

Zenvgg1X: An Advance Hybrid Framework based on Gene Sequencing for Analysis and Prediction of Lung Cancers using KNN-GNN and Custom VGG19

Zenvgg1X: An Advance Hybrid Framework based on Gene Sequenci...

引用

Communication, Computing and Signal Processing (IICCCS), IEEE International Conference on

作者： Kalpa Pandit Ahan Chatterjee Debtan Chowdhury Debraj Chatterjee Dept of Artificial Intelligence and Machine Learning Techno International New Town Kolkata India Dept of Computer Science Engineering Techno International New Town Kolkata India

ISBN: (数字)9798350390759

ISBN: (纸本)9798350390766

Lung cancer affects about 238,340 people annually, of which men account for 117,550 cases whereas women have 120,790. Lung cancer causes roughly 127,070 deaths every year. The survival rate for lung cancer is only 56 percent if it can be detected within the lungs; however, it drops to just 5 percentage where there are distant tumors. As few as 16 percentage of cases can be diagnosed if the cancer is identified early. The study created a new lung cancer detection framework, Zenvgg1X a diagnosis and predictive framework combining VGG19, which is a deep convolutional neural network (CNN), and K-Nearest Neighbors - Graph Neural Network (KNN-GNN) hybrid. The combination of these two allows analyzing both images and texts, thus increasing the success rate of lung cancer detection. By working with medical images for identification purposes only the VGG19 model that is well recognized in terms of image classification analysis for lung cancer while KNN-GNN determine patterns from text related to diseases-which if you ask me, the combinations sometimes provide statistical outputs that are mindbogglingly high in accuracy such as 87.5 percent using only 30 percent of the data or more impressively reaching up to an astonishing figure around 98 percent with larger dataset. Further inclusion of Tp53 gene mutation data can enable early-stage prediction of lung cancer which may lead to improved survival rates. It goes without saying that this hybrid model represents a giant stride in medical diagnostics because it offers hope for what could eventually become a simple routine test for early diagnosis and treatment of lung cancer.

关键词： Analytical models Accuracy Text recognition Lungs Lung cancer Nearest neighbor methods Data models Convolutional neural networks Tumors Image classification

来源：评论

学校读者我要写书评

暂无评论

machine learning-Enhanced Self-Management for Energy-Effective and Secure Statistics Assortment in Unattended WSNs

引用

SN computer science 2025年第2期6卷 1-10页

作者： Biswas, Preesat Mishra, Aishwarya Dixit, Rinku Sharma Dwivedi, Abhishek Choudhary, Shailee Lohmor Tiwari, Shweta Chakravarthi, M. Ayyappa Electronics & Telecommunications Government Engineering College C. G Jagdalpur India Department of Computer Science & Engineering IES College of Technology M.P Bhopal India Department of Artificial Intelligence and Machine Learning New Delhi Institute of Management Delhi India Department of Data Science and Engineering School of AIML IoT & IS CCE DS and Computer Applications Manipal University Jaipur Jaipur Rajasthan India Department of Computer Science and Engineering Gyan Ganga Institute of Technology and Sciences M.P Jabalpur India Department of CSE-Data Science KKR and KSR Institute of Technology and Sciences Guntur India

Unattended Wireless Sensor Networks (UWSNs) operate without human supervision and have limited resources. In a conventional WSN with a fixed sink for data collection, nodes one hop away from the sink consume more energy, causing a hotspot issue. This leads to network fragmentation and reduced sensing efficiency. To address this, a mobile sink is proposed. This study details the development and implementation of Secure MRL, focusing on legitimate message communication to the mobile sink via convex nodes. By using these convex nodes as data collection points, the mobile sink avoids traversing the entire network. We employ an enhanced system to generate keys and authenticate messages, assessing its resilience against various attacks, including sleep deprivation, snooze, network substitution, and modification. Findings indicate that Secure MRL maintains consistent performance across different node counts and significantly reduces energy consumption by 84%, improving the speed of attack detection. This improvement is achieved by monitoring the digital signature and residual energy of nodes. The proposed SVM-based network training establishes a residual energy cut-off, while convex nodes mitigate synchronization issues among other sensor nodes in Secure MRL. Additionally, convex nodes manage the availability checks for the mobile sink. The proposed approach demonstrates a 92% improvement in packet delivery ratio (PDR) and an 89% reduction in end-to-end delay, highlighting its efficacy in enhancing network performance. © The Author(s), under exclusive licence to Springer Nature Singapore Pte Ltd. 2025.

关键词： Duty-cycled UWSN MRL Sec-MRL SEC-TMP SVM UWSN

来源：评论

学校读者我要写书评

暂无评论

Comparative Analysis of Custom Neural Network Model and VGG-16 for Multi-class Image Classification

Comparative Analysis of Custom Neural Network Model and VGG-...

引用

Emerging Smart Computing and Informatics (ESCI), Conference on

作者： Khalid Saifullah Sufiyan Chaudhary Mohammed Danyaal Shaikh Muzzammil Amrelia Chaitali Mahajan Computer Science (Artificial Inteligence And Machine Learning) M.H Saboo Siddik College of Engineering Mumbai India

ISBN: (数字)9798331515683

ISBN: (纸本)9798331515690

This study assesses the performance of CustomNet, a lightweight neural network model trained using NumPy and Pandas, compared to the VGG-16 architecture on the datasets of MNIST, Fashion MNIST, and Virus MNIST. With fewer parameters, CustomNet achieves competitive accuracy since it is designed for reduced computational complexity. Model performance was assessed using metrics such F1 score, accuracy, precision, and recall. CustomNet provides a compelling trade-off between computational efficiency and accuracy, making it appropriate for resource-constrained contexts, even if the VGG-16 model showed better accuracy.

关键词： Performance evaluation Analytical models Accuracy computer viruses Computational modeling Neural networks computer architecture Mobile handsets Viruses (medical) Image classification

来源：评论

学校读者我要写书评

暂无评论

Enhanced Model for Mango Detection and Quality Classification Using Optimized Feature Extraction Techniques

Enhanced Model for Mango Detection and Quality Classificatio...

引用

IEEE Students' Conference on Electrical, Electronics and computer science (SCEECS)

作者： Adla Aryan Abdul Aleem Mohammed Manish Chabra Syed Saarib Rasheed Mohammed Adnan Mohammed Abdul Raoof Department of Artificial Intelligence & Machine Learning Vardhaman College of Engineering Hyderabad Telangana India Department of Computer Science and Engineering Muffakham Jah College of Engineering and Technology Hyderabad India Department of Artificial Intelligence & Data Science Methodist College of Engineering and Technology Hyderabad Telangana India

ISBN: (数字)9798331529833

ISBN: (纸本)9798331529840

This paper introduces an automated grading system for mangoes, enhancing efficiency and accuracy compared to human-based methods. The system uses the Lion Assisted Firefly Algorithm (LA-FF) to extract the best features from multiple highlights, enhancing grading efficiency and accuracy. The LA-FF algorithm is then used to fine-tune the convolutional layers of a deep CNN based on the specific requirements of mango grading. The system integrates the latest algorithms, automation, and adaptation to create an even more effective and precise grading system suitable for rural agricultural contexts. The LA-FF algorithm is used to extract the best features from multiple highlights, resulting in a more accurate and efficient grading process.

关键词： Measurement computer science Accuracy Feature extraction Classification algorithms Convolutional neural networks Reliability Optimization Consumer electronics Convergence

来源：评论

学校读者我要写书评

暂无评论

ClinicCare Manager: An Efficient Electronic Health Record (EHR) and Healthcare Management System Using Spring Boot and React

ClinicCare Manager: An Efficient Electronic Health Record (E...

引用

IoT Based Control Networks and Intelligent Systems (ICICNIS), International Conference on

作者： Bindu S Ashok K Keerthana Krishnamurthy Chethan C Gajula Srinivas Jatinkrishna V Gagan Department of Artificial Intelligence and Machine Learning New Horizon College of Engineering Bengaluru India Department of Computer Science & Engineering New Horizon College of Engineering Bengaluru India

ISBN: (数字)9798331518097

ISBN: (纸本)9798331518103

Healthcare institutions often face challenges in managing patient data, optimizing clinic workflows, and ensuring timely reporting. Smaller clinics, in particular, may lack the resources to streamline services for both patients and healthcare providers. This work introduces a platform that automates patient data analysis and report generation to address these needs. Built using React, Spring Boot, and PostgreSQL, the application provides efficient, end-to-end service with robust data handling and fast processing. Performance metrics show efficient resource utilization, with CPU usage rarely exceeding 3 % and heap memory remaining well within the 2048 MB limit. Compared to other health care applications, this platform offers greater scalability and responsiveness. Additionally, the framework supports secure authentication and role-based access, making it a reliable and scalable solution that meets the expanding demands of health care environments.

关键词： Data handling Scalability Authentication Medical services Regulation Real-time systems Resource management Reliability Springs Predictive analytics

来源：评论

学校读者我要写书评

暂无评论

Review on Event Extraction for BioNLP with a Survey

Review on Event Extraction for BioNLP with a Survey

引用

Advancement in Technology (ICONAT), International Conference for

作者： Veena V Pattankar P Priyanga Computer Science and Engineering Global Academy of Technology Bangalore India Artificial Intelligence and Machine Learning Global Academy of Technology Bangalore India

The scientific literature contains essential information connected to proteins, drugs, and symptoms. Researchers are extracting organized, brief, and understandable information from unstructured texts to keep up with the growing number of publications. Within the context of the biomedical area, this study investigates event extraction and natural language comprehension. We will build a flexible description of an event by first outlining several terminological methods. Second, we demonstrate event extraction, its challenges, and annotated corpora. Third, we investigate representative methodologies and provide an up-to-date analysis with performance discussion. We provide a taxonomy to assist academics in navigating the deluge of event extraction works. Fourth, we compare biomedical solutions to those in other domains to uncover fresh research opportunities and approaches. Finally, we discuss potential applications, explainability, and knowledge injection.

关键词： Proteins Drugs Navigation Biological system modeling Taxonomy Semantics Natural languages

来源：评论

学校读者我要写书评

暂无评论

Weakly Supervised Test-Time Domain Adaptation for Object Detection

arXiv

引用

arXiv 2024年

作者： Doan, Anh-Dzung Nguyen, Bach Long Lim, Terry Jayawardhana, Madhuka Gupta, Surabhi Guettier, Christophe Reid, Ian Wagner, Markus Chin, Tat-Jun Australian Institute for Machine Learning The University of Adelaide Australia Department of Data Science and Artificial Intelligence Monash University Australia Safran

Prior to deployment, an object detector is trained on a dataset compiled from a previous data collection campaign. However, the environment in which the object detector is deployed will invariably evolve, particularly in outdoor settings where changes in lighting, weather and seasons will significantly affect the appearance of the scene and target objects. It is almost impossible for all potential scenarios that the object detector may come across to be present in a finite training dataset. This necessitates continuous updates to the object detector to maintain satisfactory performance. Test-time domain adaptation techniques enable machine learning models to self-adapt based on the distributions of the testing data. However, existing methods mainly focus on fully automated adaptation, which makes sense for applications such as self-driving cars. Despite the prevalence of fully automated approaches, in some applications such as surveillance, there is usually a human operator overseeing the system’s operation. We propose to involve the operator in test-time domain adaptation to raise the performance of object detection beyond what is achievable by fully automated adaptation. To reduce manual effort, the proposed method only requires the operator to provide weak labels, which are then used to guide the adaptation process. Furthermore, the proposed method can be performed in a streaming setting, where each online sample is observed only once. We show that the proposed method outperforms existing works, demonstrating a great benefit of human-in-the-loop test-time domain adaptation. Our code is publicly available at https://***/dzungdoan6/WSTTA Copyright © 2024, The Authors. All rights reserved.

关键词： Object detection

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：