检索结果-内蒙古大学图书馆

2022 IEEE International Conference on Big data, Big data 2022

作者： Kan, Xuan Kong, Yunchuan Yu, Tianwei Guo, Ying Emory University Department of Computer Science GA United States Emory University Department of Biostatistics and Bioinformatics GA United States The Chinese University of HongKong School of Data Science Shenzhen China

ISBN: (纸本)9781665480451

Multimodal brain networks extracted from functional magnetic resonance imaging (fMRI) characterize complex connectivities among brain regions from both structural and functional, showing great potential for mental health analysis. Deep neural network models have led a tremendous success in various downstream tasks. However, common property in brain network data, the small number of samples compared with the huge amount of features, hinders the application of deep learning techniques for brain network analysis. This work presents a graph-embedding method by leveraging the unique characteristics of brain networks to unleash the power of deep neural networks and achieve outstanding prediction performance with proper explainability. Experiment results show clear advancements in our proposed braceNet on both real and synthetic datasets. © 2022 IEEE.

关键词： Graph neural networks

来源：评论

学校读者我要写书评

暂无评论

Uncertainty quantification over graph with conformalized graph neural networks 23

Uncertainty quantification over graph with conformalized gra...

引用

Proceedings of the 37th International Conference on Neural Information Processing Systems

作者： Kexin Huang Ying Jin Emmanuel Candès Jure Leskovec Department of Computer Science Stanford University Department of Statistics Stanford University Department of Statistics Stanford University and Department of Mathematics Stanford University

Graph Neural Networks (GNNs) are powerful machine learning prediction models on graph-structured data. However, GNNs lack rigorous uncertainty estimates, limiting their reliable deployment in settings where the cost of errors is significant. We propose conformalized GNN (CF-GNN), extending conformal prediction (CP) to graph-based models for guaranteed uncertainty estimates. Given an entity in the graph, CF-GNN produces a prediction set/interval that provably contains the true label with pre-defined coverage probability (e.g. 90%). We establish a permutation invariance condition that enables the validity of CP on graph data and provide an exact characterization of the test-time coverage. Besides valid coverage, it is crucial to reduce the prediction set size/interval length for practical use. We observe a key connection between non-conformity scores and network structures, which motivates us to develop a topology-aware output correction model that learns to update the prediction and produces more efficient prediction sets/intervals. Extensive experiments show that CF-GNN achieves any pre-defined target marginal coverage while significantly reducing the prediction set/interval size by up to 74% over the baselines. It also empirically achieves satisfactory conditional coverage over various raw and network features.

关键词：

来源：评论

学校读者我要写书评

暂无评论

Cognizant Care Machine Learning-Assisted Smart Records for Improved Healthcare 2

Cognizant Care Machine Learning-Assisted Smart Records for I...

引用

2nd IEEE International Conference on Integrated Circuits and Communication Systems, ICICACS 2024

作者： Mageshwari, G. Latha, J. Naveen, G. Murugan, S. Karthick Joseph Gilbert Andrews, Allen R.M.K.College of Engineering and Technology Department of Artificial Intelligence and Data Science Tiruvallur Dist Puduvayal India R.M.K.College of Engineering and Technology Department of Computer Science and Engineering Tiruvallur Dist Kavaraipettai India Madha Institute of Engineering and Technology Department of Electronics and Communication Engineering Tiruvallur Dist Kavaraipettai India R. M. D. Engineering College Department of Computer Science and Engineering Tiruvallur Dist Kavaraipettai India Deloitte Consulting LLP 200 Sterling Road MechanicsburgPA United States

ISBN: (纸本)9798350317558

In the era of evolving healthcare demands, the requirement for innovation in record-keeping systems is undeniable. Traditional methods of documenting have difficulties in terms of efficiency, accuracy, and flexibility. The paper addresses an urgent need for a revolutionary solution by proposing "Cognizant Care,"a unique system that employs machine learning to build Smart Records, therefore revolutionizing healthcare documentation and decision-making processes. The inefficiencies of legacy systems, which include time-consuming manual data entry and limited predictive capabilities, highlight the need for a novel concept. Cognizant Care's architecture can be incorporated with existing Electronic Health Record systems, providing intelligent data extraction, predictive analytics, decision support, and automated reporting. The results and analysis show a significant improvement demonstrates superior performance with accuracy at 94%, precision of 92%, recall of 88%, and an F1 score of 90%, proving the system's ability to mitigate the drawbacks of existing systems. Cognizant Care emerges as an essential method signifying an evolving era in which the combination of human knowledge and artificial intelligence drives healthcare into a realm of unparalleled achievements. © 2024 IEEE.

关键词： Cognizant Care Cost Savings Decision Support Electronic Health Records Healthcare Innovation Personalized Patient Care

来源：评论

学校读者我要写书评

暂无评论

Sharp Matrix Empirical Bernstein Inequalities

arXiv

引用

arXiv 2024年

作者： Wang, Hongjian Ramdas, Aaditya Department of Statistics and Data Science Carnegie Mellon University United States Machine Learning Department Carnegie Mellon University United States

We present two sharp empirical Bernstein inequalities for symmetric random matrices with bounded eigenvalues. By sharp, we mean that both inequalities adapt to the unknown variance in a tight manner: the deviation captured by the first-order 1/√n term asymptotically matches the matrix Bernstein inequality exactly, including constants, the latter requiring knowledge of the variance. Our first inequality holds for the sample mean of independent matrices, and our second inequality holds for a mean estimator under martingale dependence at stopping times. Copyright © 2024, The Authors. All rights reserved.

关键词： Matrix algebra

来源：评论

学校读者我要写书评

暂无评论

A Lightweight Encryption Method for Preserving E-Healthcare data Privacy Using Dual Signature on Twisted Edwards Curves 1

引用

5th International Conference on computer and Communication Technologies, IC3T 2023

作者： Farhat, Suryya Kumar, Manoj Vaish, Arti Dewangan, Bhupesh Kumar Choudhury, Tanupriya Kotecha, Ketan Department of Applied Science and Humanities ADGITM Delhi110053 India Department of Mathematics and Statistics Gurukula Kangri Vishwavidyalaya Haridwar249404 India Department of Computer Science and Engineering OP Jindal University Raigarh496109 India Lavale Campus Maharashtra Pune412115 India Pune412115 India Research Professor CSE Dept. Graphic Era Deemed to be University Uttarakhand Dehradun248002 India

ISBN: (数字)9789819997077

ISBN: (纸本)9789819997060

Due to covid-19 pandemic, research in e-healthcare system is gaining popularity because in most of the cases e-healthcare system does not require to present patient physically at doctor’s door. The reason behind this is that e-healthcare system eliminates the necessity for patient to travel at healthcare centres for doctor’s prescription or to collect their diseases diagnostic reports. In the present paper, we have proposed an algorithm, which may be capable to provide integrity for e-healthcare data and preserve its privacy also. The proposed scheme combines dual signature with EdDSA. Dual signature looks similar to double signature but it is quite different;it is a technique of coupling to different values of varied natures preserving them anonymous to varied objects securely. From recent research, it has been proved that elliptic curve cryptography is more suitable for resource constraint (lightweight) device when compared with RSA. Being an improvement of ECDSA, EdDSA signature generation is egregious and affordable. Moreover, EdDSA signature verification, being computationally very hard task, cannot be carried out by e-healthcare devices. Although edge computing devices having no operational constraints can easily perform this task. These characteristics make our proposal more elegant, perky, and efficient strategy for e-healthcare system. © The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd. 2024.

关键词： Diagnosis

来源：评论

学校读者我要写书评

暂无评论

A New Extension of the Topp–Leone-Family of Models with Applications to Real data

引用

Annals of data science 2023年第1期10卷 225-250页

作者： Muhammad, Mustapha Liu, Lixia Abba, Badamasi Muhammad, Isyaku Bouchane, Mouna Zhang, Hexin Musa, Sani Department of Mathematics Guangdong University of Petrochemical Technology Maoming525000 China School of Mathematical Sciences Hebei Normal University Shijiazhuang050024 China School of Mathematics and Statistics Central South University Changsha China Department of Mathematics Yusuf Maitama Sule University Kano Nigeria School of Mechanical and Electrical Engineering University of Electronic Science and Technology of China Chengdu611731 China Key Laboratory of Augmented Reality College of Mathematics and Information Science Hebei Normal University Shijiazhuang China Department of Mathematics and Computer Science Sule Lamido University Jigawa Kafin-Hausa Nigeria

In this article, we proposed a new extension of the Topp–Leone family of distributions. Some important properties of the model are developed, such as quantile function, stochastic ordering, model series representation, moments, stress–strength reliability parameter, Renyi entropy, order statistics, and moment of residual life. A particular member called new extended Topp–Leone exponential (NETLE) is discussed. Maximum likelihood estimation (MLE), least-square estimation (LSE), and percentile estimation (PE) are used for the model parameter estimation. Simulation studies were conducted using NETLE to assess the MLE, LSE, and PE performance by examining their bias and mean square error (MSE), and the result was satisfactory. Finally, the applications of the NETLE to two real data sets are provided to illustrate the importance of the NETLG families in practice;the data sets consist of daily new deaths due to COVID-19 in California and New Jersey, USA. The new model outperformed many other existing Topp–Leone’s and exponential related distributions based on the real data illustrations. © 2022, The Author(s), under exclusive licence to Springer-Verlag GmbH Germany, part of Springer Nature.

关键词： Maximum likelihood estimation

来源：评论

学校读者我要写书评

暂无评论

Maize Plant Disease Prediction of UAV Images for Precision Agriculture Using Fusion of Multimodal 3

Maize Plant Disease Prediction of UAV Images for Precision A...

引用

3rd International Conference on Computing and Information Technology, ICCIT 2023

作者： Mishra, Vinay Naik, Nenavath Srinivas Kumar, Santosh Alsamhi, Saeed Hamood Saif, Abdu Curry, Edward -Naya Raipur Department of Computer Science and Engineering Chhattisgarh India University of Galway Insight Centre for Data Analytics Galway Ireland Faculty of Engineering & Information Technology Taiz Yemen

ISBN: (纸本)9798350321487

The agriculture sector is vital to a country's economic and productive development. With technological advancements, we can identify plant diseases in their early stages, similar to how human diseases are diagnosed, and avoid significant financial loss. For example, a new technique has been developed to detect diseases in maize crops, a widely used raw product in the food industry and precision agriculture. Furthermore, farmers ensure the fields do not remain wet for extended periods to prevent fungus infection. Currently, machine learning and deep learning approaches are being utilized, making agriculture one of the most researched topics in AI. This study proposes a novel framework for predicting maize plant disease using a fusion of multi-deep learning models. The proposed framework uses a four-stage pipeline for detecting disease in maize plants by fusion of Xception, DenseNet-121, and ensembling of Xception and DenseNet-121 models to achieve excellent results on the OSF drone image dataset, which consists of 1821 high-resolution images of maize plants taken from unmanned aerial vehicles (UAVs). The accuracy of our proposed approach is 96.95%, which is highly effective, considering the challenges posed by drone images. © 2023 IEEE.

关键词： Drones

来源：评论

学校读者我要写书评

暂无评论

HAR-CTGAN: A Mobile Sensor data Generation Tool for Human Activity Recognition

HAR-CTGAN: A Mobile Sensor Data Generation Tool for Human Ac...

引用

2022 IEEE International Conference on Big data, Big data 2022

作者： Deoliveira, Joshua Gerych, Walter Koshkarova, Aruzhan Rundensteiner, Elke Agu, Emmanuel Data Science Program Worcester Polytechnic Institute WorcesterMA United States Department of Computer Science Worcester Polytechnic Institute WorcesterMA United States

ISBN: (纸本)9781665480451

Human activity recognition (HAR) is the process of using mobile sensor data to determine the physical activities performed by individuals. HAR is the backbone of many mobile healthcare applications, such as passive health monitoring systems, early diagnosing systems, and fall detection systems. Effective HAR models rely on deep learning architectures and big data in order to accurately classify activities. Unfortunately, HAR datasets are expensive to collect, are often mislabeled, and have large class imbalances. State-of-the-art approaches to address these challenges utilize Generative Adversarial Networks (GANs) for generating additional synthetic data along with their labels. Problematically, these HAR GANs only synthesize continuous features - features that are represented by real numbers - recorded from gyroscopes, accelerometers, and other sensors that produce continuous data. This is limiting since mobile sensor data commonly has discrete features that provide additional context such as device location and the time-of-day, which have been shown to substantially improve HAR classification. Hence, we studied Conditional Tabular Generative Adversarial Networks (CTGANs) for data generation to synthesize mobile sensor data containing both continuous and discrete features, a task never been done by state-of-the-art approaches. We show HAR-CTGANs generate data with greater realism resulting in allowing better downstream performance in HAR models, and when state-of-the-art models were modified with HAR-CTGAN characteristics, downstream performance also improves. © 2022 IEEE.

关键词： Large dataset

来源：评论

学校读者我要写书评

暂无评论

Exploring LLM and Neural Networks Towards Malicious Prompt Detection

Exploring LLM and Neural Networks Towards Malicious Prompt D...

引用

Electronics and Sustainable Communication Systems (ICESC), 2020 International Conference on

作者： Himani Deshpande Dhawal Chaudhari Tanmay Sarode Ayush Kamath Department of Artificial Intelligence & Data Science Thadomal Shahani Engineering College Mumbai India Department of Computer Science Thadomal Shahani Engineering College Mumbai India Department of Computer Science Thadomal Shahani Engineering College Mumbai India

ISBN: (数字)9798350379945

ISBN: (纸本)9798350379952

Atthe forefront of the Artificial Intelligence Revolution is the Generative AI domain which is making splashes in generation of new content from existing Large Language Models. Large Language Models (LLMs) are flexible, effective tools with many uses. On the other hand, they can be tricked by devious provocation. This work explores the detection of fraudulent prompts intended to produce false or damaging outputs using Long Short-Term Memory (LSTM) networks. During this study we analysed the LTSM model against a neural network model. We found a 91.69% accuracy for the LTSM model and a 92.52% accuracy for the simple neural network. However, the simple neural network had a higher F1 score of 0.73 compared to the LTSM score of 0.69. We found that the standard neural model is more efficient at identifying harmful user prompts. Our research highlights the need for taking into account various architectures of models to reduce the risk of malicious prompting in language learning models.

关键词： Training Measurement Analytical models Accuracy Computational modeling Large language models Neural networks Security Long short term memory Standards

来源：评论

学校读者我要写书评

暂无评论

ZeroMark: towards dataset ownership verification without disclosing watermarks 24

ZeroMark: towards dataset ownership verification without dis...

引用

Proceedings of the 38th International Conference on Neural Information Processing Systems

作者： Junfeng Guo Yiming Li Ruibo Chen Yihan Wu Chenxi Liu Heng Huang Department of Computer Science Institute of Health Computing University of Maryland College Park College of Computing and Data Science Nanyang Technology University

ISBN: (纸本)9798331314385

High-quality public datasets significantly prompt the prosperity of deep neural networks (DNNs). Currently, dataset ownership verification (DOV), which consists of dataset watermarking and ownership verification, is the only feasible solution to protect their copyright by preventing unauthorized use. In this paper, we revisit existing DOV methods and find that they all mainly focused on the first stage by designing different types of dataset watermarks and directly exploiting watermarked samples as the verification samples for ownership verification. As such, their success relies on an underlying assumption that verification is a onetime and privacy-preserving process, which does not necessarily hold in practice. To alleviate this problem, we propose ZeroMark to conduct ownership verification without disclosing dataset-specified watermarks. Our method is inspired by our empirical and theoretical findings of the intrinsic property of DNNs trained on the watermarked dataset. Specifically, ZeroMark first generates the closest boundary version of given benign samples and calculates their boundary gradients under the label-only black-box setting. After that, it examines whether the given suspicious method has been trained on the protected dataset by performing a hypothesis test, based on the cosine similarity measured on the boundary gradients and the watermark pattern. Extensive experiments on benchmark datasets verify the effectiveness of our ZeroMark and its resistance to potential adaptive attacks. The codes for reproducing our main experiments are publicly available at https://***/JunfengGo/***.

关键词：

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：