检索结果-内蒙古大学图书馆

Biomedical Materials and Devices 2025年 1-22页

作者： Chowdhury, Abir Druvo, Md Mahbubur Rahman Uddin, Khandaker Mohammad Mohi Islam, Md.Shariful Uddin, Md Ashraf Department of Computer Science and Engineering Dhaka International University Dhaka 1205 Bangladesh Department of Computer Science and Engineering Southeast University Dhaka International University Dhaka 1208 Bangladesh School of Information and Technology Deakin University Geelong Warun Ponds Australia

Celiac disease is a chronic autoimmune illness caused by gluten consumption that affects millions worldwide. Accurate diagnosis is critical for optimal care but traditional approaches frequently lack precision. The proposed work uses machine learning approaches to improve diagnostic accuracy on a dataset that includes clinical and laboratory variables. Preprocessing the data involved firstly scaling and encoding the categorical values. Then, selecting the important features using the recursive feature elimination (RFE) with an extra tree classifier (ETC) (RFE + ETC) and assessing the five machine learning models KNN, GNB, MNB, BNB, and QDA were important procedures. KNN outperformed the others, achieving 99.49% accuracy, 99.36% precision, 100% recall, and an AUC of 99.98% on the selected feature. Furthermore, the study integrated the findings into a web application to aid in real-time diagnostic decision-making, assuring clinical applicability. This study emphasizes the necessity of strong feature selection and advanced machine learning approaches in improving celiac disease diagnosis and therapy. Shapley additive explanations (SHAP) were finally incorporated for interpretability, providing details on different variables affect model predictions. © The Author(s), under exclusive licence to Springer science+Business Media, LLC, part of Springer Nature 2025.

关键词： Celiac disease Interpretability KNN Machine learning RFE + ETC SHAP Web application

来源：评论

学校读者我要写书评

暂无评论

Improved method of dialogue model based on character and word fusion 2

Improved method of dialogue model based on character and wor...

引用

2nd International Conference on Electronic Information Technology, EIT 2023

作者： Yang, Qingli Wang, Zhiliang Xiao, Ruoxiu School of Computer and Communication Engineering University of Science and Technology Beijing Shunde Innovation School Foshan China School of Computer and Communication Engineering University of Science and Technology Beijing Beijing China

ISBN: (数字)9781510666627

ISBN: (纸本)9781510666610

An important part of the dialogue robot is the understanding of semantics and the analysis of sentences. At present, most of the models are trained based on character vector and word vector. Compared with traditional models, this method can fully consider the long and irregular language characteristics of daily chit-chat dialogues, and make full use of character and word fusion to improve the dialogue performance of the model. At the same time, the word embedding matrix trained by the pre-trained model is calculated to obtain the same dimension as the character embedding. Then, the character vector and word vector are fused, and the transformer model is used for dialogue training. The experimental results show that the proposed method can better improve the understanding ability of the dialogue model. © 2023 SPIE.

关键词： Semantics

来源：评论

学校读者我要写书评

暂无评论

Biometric-Based Key Handling Using Variation to Scale Invariant Feature Transform amid COVID-19 Pandemic 7th

Biometric-Based Key Handling Using Variation to Scale Invar...

引用

7th International Conference on Advanced Computing and Intelligent engineering, ICACIE 2022

作者： Kaur, Prabhjot Kumar, Nitin Singh, Maheep Department of Computer Science and Engineering National Institute of Technology Uttarakhand Uttarakhand Srinagar India School of Computing DIT University Uttarakhand Dehradun India Department of Computer Science and Engineering PEC Deemed-to-be University Chandigarh India Department of Computer Science and Engineering Doon University Uttarakhand Dehradun India

ISBN: (纸本)9789819950140

The outbreak of COVID-19 pandemic has hardened the biometric recognition, authentication, and cryptographic process based on facial features due to masks coverage. The Biometric cryptosystem (BCS) consisting of touch sensing biometric, such as fingerprint, palmprint, vein, is no longer in use in this scenario. In this context, other key-based biometric cryptosystems based on facial biometric may underperform. Thus, there is a need to simulate existing and implement novel biometric cryptosystems based on other biometric cues such as Ear, Face, Iris, Gait. This paper introduces a novel method for biometric key generation based on Scale Invariant Feature Transform (SIFT). The obtained key is used to encrypt/decrypt the data. At enrollment stage, the biometric key is obtained which subsequently encrypts the data and authentication phase decrypts the secret upon successful biometric match. The proposed method is compared with the state-of-art methods on three datasets namely AMI (ear), FACES (face), and UBIPr (iris). The metrics used to evaluate the performance of proposed method include: NPCR, UACI, PSNR, Correlation, MSE, MAE, SSIM, NRMSE. © The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd. 2024.

关键词： Biometrics

来源：评论

学校读者我要写书评

暂无评论

Delay-Constrained Multicast Throughput Maximization in MEC Networks for High-Speed Railways 19th

Delay-Constrained Multicast Throughput Maximization in MEC ...

引用

19th EAI International Conference on Collaborative Computing: Networking, Applications and Worksharing, CollaborateCom 2023

作者： Xu, Junyi Wei, Zhenchun Yuan, Xiaohui Lyu, Zengwei Feng, Lin Han, Jianghong School of Computer Science and Information Technology Hefei University of Technology Hefei China Department of Computer Science and Engineering University of North Texus DentonTX United States

ISBN: (纸本)9783031545306

Multi-access Edge Computing presents a compelling solution for delivering seamless connectivity to computing services. In this study, we aim to optimize multicast throughput to ensure high-quality experiences for passengers engaged in inter-train interactions within dedicated MEC networks designed for high-speed railways. Considering the unique challenges associated with high-speed railways, we model multicast routing paths as group Steiner trees. Subsequently, we devise a rapid tree construction method by converting the root search into the Generalized Assignment Problem (GAP). This innovative approach skillfully balances accuracy and computational efficiency. We demonstrate that this problem can be effectively reduced to the bounded knapsack problem with setups. In addition, we recognize the presence of precedence constraints between tasks and their respective outcomes. Consequently, we introduce a new variant of the knapsack problem, which we refer to as the Precedence-constrained Bounded Knapsack Problem with Setups. Our approach, termed the GAP- and knapsack-based Group Steiner Tree (GKGST), offers a relative performance guarantee of 1/2. We evaluate the GKGST algorithm against three baseline algorithms, which are adapted and extended from existing literature. Simulation results indicate that our proposed algorithm exhibits considerable potential for enhanced performance. © ICST Institute for computer sciences, Social Informatics and Telecommunications engineering 2024.

关键词： Computational efficiency

来源：评论

学校读者我要写书评

暂无评论

Investigation of Single Channel Speech Enhancement: A Comparative Study 4

Investigation of Single Channel Speech Enhancement: A Compar...

引用

4th International Conference on Neural Networks, Information and Communication engineering, NNICE 2024

作者： Ma, Songqi Lv, Zhen Du, Fei Zhang, Xiao University of Science and Technology Beijing School of Computer and Communication Engineering Beijing China University of Science and Technology Beijing School of Advance Engineering Beijing China

ISBN: (纸本)9798350394375

Noise is a critical factor that impacts speech quality. Speech signals are often disturbed by noise during acquisition and transmission. Speech enhancement technology represents an important method to eliminate noise from noisy signal and extract clean speech, which holds significant value in the field of speech recognition, speech communication. This study first categorizes the prior characteristics of different types of noise, followed by an exposition and implementation of representative speech enhancement methods, including traditional signal processing methods and deep learning algorithms. Through three metrics of evaluation, the performance of various speech enhancement algorithms in distinct noise environments is compared, and the Entropy Weight-Topsis (EW-Topsis) method is used for synthesis. Finally, the applicability and development trends of existing speech enhancement algorithms were discussed. The innovation of this study lies in the comparison of different algorithms in various noisy environments using different evaluation metrics. Previous research has mainly focused on two of the evaluation, algorithms and noise. This study presents a comprehensive comparison to synthesize three aspects, providing a unified evaluation reference. © 2024 IEEE.

关键词： Speech enhancement

来源：评论

学校读者我要写书评

暂无评论

DNN-based Secure Remote Patient Data Analysis Framework for Improving Human Life Expectancy in Healthcare 4.0 2

DNN-based Secure Remote Patient Data Analysis Framework for ...

引用

2nd International Conference on Machine Learning and Data engineering, ICMLDE 2023

作者： Darji, Krisha Ramoliya, Fenil Kakkar, Riya Gupta, Rajesh Tanwar, Sudeep Garg, Deepak Department of Computer Science and Engineering Institute of Technology Nirma University Ahmedabad India School of Computer Science and Artificial Intelligence SR University Telangana Warangal506371 India

The evolution of healthcare from its early beginnings to the recent healthcare 4.0 revolution has remarkably improved human life and living standards. The integration of emerging technologies has played a pivotal role in this progress. Notably, remote analysis of patient data has emerged as a promising approach, enabling telemedicine and remote patient monitoring enhancing healthcare accessibility and efficiency. However, the vulnerabilities of patient data to malicious network attacks pose critical challenges in the healthcare industry, potentially compromising patients' safety and undermining trust in the system. Thus, we have introduced a cutting-edge deep neural network (DNN) model to mitigate the risks associated with remote patient data transfer in healthcare 4.0 by bifurcating the data into malicious or non-malicious. The primary objective of the proposed framework is to ensure the secure and private communication of patient data, thereby fostering a more dependable and trustworthy healthcare ecosystem. Finally, the proposed framework is evaluated against various standard metrics such as accuracy, loss considering binary cross-entropy, receiver operating characteristic (ROC) curve, precision-recall curve and confusion matrix. The implications of the proposed framework offer data security to the healthcare system as it contributes to creating a more resilient and dependable ecosystem, thus promoting better patient care and ultimately elevating the overall life expectancy and well-being of individuals. © 2024 Elsevier B.V.. All rights reserved.

关键词： Deep neural networks

来源：评论

学校读者我要写书评

暂无评论

'Optimizing Algorithmic Efficiency: Exploring Advanced Data Structures for High-Performance Computing' 3

'Optimizing Algorithmic Efficiency: Exploring Advanced Data ...

引用

3rd International Conference on Computational Modelling, Simulation and Optimization, ICCMSO 2024

作者： Rao, Lade Gunakar Nitya, Erupaka Devi Kusumba, Jyoshna Rajchandar, K. Narasimha Murthy, S. SR University School of Computer Science and Artificial Intelligence Telangana Warangal India Telangana Hyderabad India Government Degree College Computer Science And Applications Telangana Warangal India Department of Computer Science and Engineering Telangana Hyderabad India

ISBN: (纸本)9798350361391

This work explores advanced data structures as a means of optimizing algorithmic efficiency in high-performance computing. The search for faster and more scalable algorithms becomes essential as computing demands rise further. This paper explores state-of-the-art data structures intended to improve overall performance, memory usage, and computational speed. We hope to offer insights into the applicability and impact of various advanced data structures on high-performance computing tasks through a thorough review and analysis of them, including but not limited to self-balancing trees, hash tables, and advanced graph structures. The results of this study not only advance our theoretical knowledge of these data structures but also provide developers and researchers with useful advice on how to utilize algorithmic efficiency in quickly fully. It is important to remember that algorithmic urban planning is an interdisciplinary field that encompasses data science, computer science, urban planning, and many other fields. The goal of integrating algorithms and smart technologies is to build more efficient, sustainable, and liveable urban environments.[8]. © 2024 IEEE.

关键词： Advanced Data Structures Algorithmic Efficiency Computing (keywords) High-Performance Computing Optimizing

来源：评论

学校读者我要写书评

暂无评论

Few-Shot Semantic Dependency Parsing via Graph Contrastive Learning 30

Few-Shot Semantic Dependency Parsing via Graph Contrastive L...

引用

Joint 30th International Conference on Computational Linguistics and 14th International Conference on Language Resources and Evaluation, LREC-COLING 2024

作者： Li, Bin Fan, Yunlong Sataer, Yikemaiti Shi, Chuanqi Gao, Miao Gao, Zhiqiang School of Computer Science and Engineering Southeast University Nanjing210096 China Key Laboratory of Computer Network and Information Integration Ministry of Education China

ISBN: (纸本)9782493814104

Graph neural networks (GNNs) have achieved promising performance on semantic dependency parsing (SDP), owing to their powerful graph representation learning ability. However, training a high-performing GNN-based model requires a large amount of labeled data and it is prone to over-fitting in the absence of sufficient labeled data. To address this drawback, we propose a syntax-guided graph contrastive learning framework to pre-train GNNs with plenty of unlabeled data and fine-tune pre-trained GNNs with few-shot labeled SDP data. Through extensive experiments conducted on the SemEval-2015 Task 18 English dataset in three formalisms (DM, PAS, and PSD), we demonstrate that our framework achieves promising results when few-shot training samples are available. Furthermore, benefiting from the pre-training process, our framework exhibits notable advantages in the out-of-domain test sets. © 2024 ELRA Language Resource Association: CC BY-NC 4.0.

关键词： Graph neural networks

来源：评论

学校读者我要写书评

暂无评论

A spatio-temporal feature learning method based on transformer

A spatio-temporal feature learning method based on transform...

引用

2024 International Conference on Signal Processing and Communication Security, ICSPCS 2024

作者： Xu, Liuquan Jiang, Xuefeng Yu, Li School of Computer Science and Engineering Anhui University of Science and Technology Huainan232001 China School of Safety Science and Engineering Anhui University of Science and Technology Huainan232001 China

ISBN: (纸本)9781510681699

With the development of the network world, people's lives are gradually becoming intelligent. However, there are also crises in the network, such as attacks by some illegal hackers through a variety of complex means. The emergence of intrusion detection systems has greatly protected the network environment. However, traditional techniques cannot sufficiently learn the characteristics of attack traffic, resulting in poor performance. In order to address the limitations of traditional models, this paper proposes a transformer-based spatio-temporal feature learning method called STF-Transformer. It studies the temporal and spatial characteristics of the attack traffic through improved encoders and decoders, respectively. It greatly enhances the detection capability of the model. To demonstrate the effectiveness of the STF-transformer method. We measured the performance of the STF-transformer method on the CICIDS2017 and NSL-KDD datasets. The experimental results demonstrated that the STF-transformer method improved the detection rate. It achieved satisfactory results in terms of accuracy, F1 score, and other metrics. © 2024 SPIE.

关键词： Intrusion detection

来源：评论

学校读者我要写书评

暂无评论

UCTransNeXt: Global and local information fusion for retinal vessel segmentation 24

UCTransNeXt: Global and local information fusion for retinal...

引用

2024 International Conference on Advanced Robotics, Automation engineering and Machine Learning, ARAEML 2024

作者： Wang, Yueru Li, Peiyuan Zhan, Qianyi Peng, Yue Jiang, Yizhang Xie, Zhenping Liu, Yuan School of Artificial Intelligence and Computer Science Jiangnan University Wuxi China College of Computer and Information Engineering Nanjing University of Technology Nanjing China

ISBN: (纸本)9798400717116

The precise segmentation of retinal blood vessels is essential for the early diagnosis of various ophthalmic diseases. Meanwhile, the U-shaped structure based on convolutional neural networks performs well in medical image segmentation. However, due to the local operation characteristics of convolution, the ability to extract global feature information is limited, making it challenging to improve segmentation accuracy on retinal vessels. This paper proposes a network model UCTransNeXt that integrates global and local information to address the above problems. First, a Transformer branch is added to the U-Net network encoder section in parallel with the original convolutional layers to model global contextual information. The BiFusion module effectively fuses this global information with the local feature information extracted by the convolutional neural network. Second, the traditional Transformer structure tends to focus on all positions globally. At the same time, the adopted Intra-Scale Transformer can better capture the local information within the same scale, resulting in higher segmentation accuracy. Also, in order to efficiently label and predict convolutional features, tokenized MLPs are used in the latent space to focus more on vessel local information. The experimental results show that the Dice coefficients of the network proposed in this article reached 90.60% and 83.84%, respectively, and the accuracy reached 98.62% and 97.25%, respectively. The proposed algorithm's overall performance surpasses existing algorithms, and the modules can effectively improve fine vessel segmentation. © 2024 ACM.

关键词： Ophthalmology

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：