检索结果-内蒙古大学图书馆

An Efficient Stacked Ensemble Model for Heart Disease Detection and Classification

computers, Materials & Continua 2023年第10期77卷 665-680页

作者： Sidra Abbas Gabriel Avelino Sampedro Shtwai Alsubai Ahmad Almadhor Tai-hoon Kim Department of Computer Science COMSATS UniversityIslamabadPakistan College of Computer Studies De La Salle UniversityManila1004Philippines Faculty of Information and Communication Studies University of the Philippines Open UniversityLos Banos4031Philippines College of Computer Engineering and Sciences Prince Sattam bin Abdulaziz UniversityAlkharjSaudi Arabia Department of Computer Engineering and Networks College of Computer and Information SciencesJouf UniversitySakaka72388Saudi Arabia School of Electrical and Computer Engineering Yeosu CampusChonnam National UniversityYeosu-si59626Korea

Cardiac disease is a chronic condition that impairs the heart’s *** includes conditions such as coronary artery disease,heart failure,arrhythmias,and valvular heart *** conditions can lead to serious complications and even be life-threatening if not detected and managed in *** have utilized Machine Learning(ML)and Deep Learning(DL)to identify heart abnormalities swiftly and *** approaches have been applied to predict and treat heart disease utilizing ML and *** paper proposes a Machine and Deep Learning-based Stacked Model(MDLSM)to predict heart disease *** approaches such as eXtreme Gradient Boosting(XGB),Random Forest(RF),Naive Bayes(NB),Decision Tree(DT),and KNearest Neighbor(KNN),along with two DL models:Deep Neural Network(DNN)and Fine Tuned Deep Neural Network(FT-DNN)are used to detect heart *** models rely on electronic medical data that increases the likelihood of correctly identifying and diagnosing heart ***-known evaluation measures(i.e.,accuracy,precision,recall,F1-score,confusion matrix,and area under the Receiver Operating Characteristic(ROC)curve)are employed to check the efficacy of the proposed *** reveal that the MDLSM achieves 94.14%prediction accuracy,which is 8.30%better than the results from the baseline experiments recommending our proposed approach for identifying and diagnosing heart disease.

关键词： Deep neural network heart disease healthcare machine learning stacking

来源：评论

学校读者我要写书评

暂无评论

Few-Shot Name Entity Recognition on StackOverflow 9

Few-Shot Name Entity Recognition on StackOverflow

引用

9th International Conference on Intelligent Computing and Signal Processing, ICSP 2024

作者： Chen, Xinwei Li, Kun Song, Tianyou Guo, Jiangjian University of Illinois at Urbana Champaign Department of Electronical and Computer Engineering UrbanaIL United States University of Illinois at Urbana Champaign Department of Computer Science UrbanaIL United States Columbia University Department of Computer Science New York United States University of California San Diego Department of Computer Science and Engineering La Jolla United States

ISBN: (纸本)9798350376548

StackOverflow, with its vast question repository and limited labeled examples, raise an annotation challenge for us. We address this gap by proposing RoBERTa+MAML, a few-shot named entity recognition (NER) method leveraging meta-learning. Our approach, evaluated on the StackOverflow NER corpus (27 entity types), achieves a 5% F1 score improvement over the baseline. We improved the results further domain-specific phrase processing enhance results. © 2024 IEEE.

关键词： Zero-shot learning

来源：评论

学校读者我要写书评

暂无评论

Mix of Experts language Model for Named Entity Recognition 6

Mix of Experts Language Model for Named Entity Recognition

引用

6th International Conference on Communications, Information System and computer engineering, CISCE 2024

ISBN: (纸本)9798350353174

Named Entity Recognition (NER) is an essential steppingstone in the field of natural language processing. Although promising performance has been achieved by various distantly supervised models, we argue that distant supervision inevitably introduces incomplete and noisy annotations, which may mislead the model training process. To address this issue, we propose a robust NER model named BOND-MoE based on Mixture of Experts (MoE). Instead of relying on a single model for NER prediction, multiple models are trained and ensembled under the Expectation-Maximization (EM) framework, so that noisy supervision can be dramatically alleviated. In addition, we introduce a fair assignment module to balance the document-model assignment process. Extensive experiments on real-world datasets show that the proposed method achieves state-of-the-art performance compared with other distantly supervised NER. © 2024 IEEE.

关键词： component Mix of expert Named Entity Recognition NLP

来源：评论

学校读者我要写书评

暂无评论

I/O Complexity of Attention, or How Optimal is FlashAttention? 41

I/O Complexity of Attention, or How Optimal is FlashAttentio...

引用

41st International Conference on Machine Learning, ICML 2024

作者： Saha, Barna Ye, Christopher Department of Computer Science and Engineering University of California La Jolla San DiegoCA United States

Attention is at the heart of the popular Transformer architecture, yet suffers from quadratic time and memory complexity. In a recent significant development, FlashAttention shows that the I/O complexity of attention is the true bottleneck in scaling Transformers. Given two levels of memory hierarchy, a fast cache (e.g. GPU on-chip SRAM) where computation happens and a slow memory (e.g. GPU high-bandwidth memory) where the data resides, the I/O complexity measures the number of accesses to the slow memory. FlashAttention is an I/O-aware algorithm for self-attention that requires NM2d2 I/O operations where N is the dimension of the attention matrix, d is the head-dimension and M is the size of cache. Naturally, to further reduce the computational costs of Attention, the authors ask the question: is FlashAttention's I/O complexity optimal for every value of M? We resolve the above question in its full generality by showing an I/O complexity lower bound that matches the upper bound provided by FlashAttention for any values of M ≥ d2 within any constant factors. Moreover, our lower bounds do not rely on using combinatorial matrix multiplication for computing the attention matrix: even if one uses fast matrix multiplication, the above I/O complexity bounds cannot be improved. Further, we give a better algorithm with lower I/O complexity for M 2, and show that it is optimal for combinatorial algorithms. We do so by introducing a new communication complexity protocol for matrix compression, and connecting communication complexity to I/O complexity. We believe this connection could be of independent interest and will find more applications in proving I/O complexity lower bounds in future. Copyright 2024 by the author(s)

关键词： computer graphics equipment

来源：评论

学校读者我要写书评

暂无评论

Inverse Reinforcement Learning with Graph Neural Networks for IoT Resource Allocation 48

Inverse Reinforcement Learning with Graph Neural Networks fo...

引用

48th IEEE International Conference on Acoustics, Speech and Signal Processing, ICASSP 2023

作者： Wang, Guangchen Cheng, Peng Chen, Zhuo Xiang, Wei Vucetic, Branka Li, Yonghui The University of Sydney School of Electrical and Information Engineering Australia Data 61 Csiro Australia La Trobe University Department of Computer Science and Information Technology Australia

ISBN: (纸本)9781728163277

The rapid development of Internet of Things (IoT) applications requires efficient computing and communication resource allocation strategies to streamline the existing network operations. These strategies could be formulated as mixed-integer nonlinear programming (MINLP) problems, where the optimal branch-and-bound (B&B) with the full strong branching (FSB) variable selection policy features an extremely high complexity. We propose inverse reinforcement learning with graph neural networks (GNNIRL) to generate a new variable selection policy that closely matches the FSB variable selection. Without sacrificing the optimality, the GNNIRL can directly infer the variable selection with a significantly lower complexity, which is also verified by simulation. © 2023 IEEE.

关键词： Internet of things

来源：评论

学校读者我要写书评

暂无评论

Genomic Mutations Within the Host Microbiome: Adaptive Evolution or Purifying Selection

引用

engineering 2023年第1期20卷 96-102页

作者： Jiachao Zhang Rob Knight School of Food Science and Engineering Hainan UniversityHaikou 570228China Department of Pediatrics University of California San DiegoLa JollaCA 92093USA Center for Microbiome Innovation Jacobs School of EngineeringUniversity of California San DiegoLa JollaCA 92093USA Department of Bioengineering University of California San DiegoLa JollaCA 92093USA Department of Computer Science and Engineering University of California San DiegoLa JollaCA 92093USA

Next-generation sequencing technology has transformed our ability to assess the taxonomic composition functions of host-associated microbiota and microbiomes. More human microbiome research projects—particularly those that explore genomic mutations within the microbiome—will be launched in the next decade. This review focuses on the coevolution of microbes within a microbiome, which shapes strain-level diversity both within and between host species. We also explore the correlation between microbial genomic mutations and common metabolic diseases, and the adaptive evolution of pathogens and probiotics during invasion and colonization. Finally, we discuss advances in methods and algorithms for annotating and analyzing microbial genomic mutations.

关键词： Gut microbiota Genomic mutations Adaptive evolution Purifying selection Single-nucleoti de variants

来源：评论

学校读者我要写书评

暂无评论

Addressing the Productivity Paradox in Healthcare with Retrieval Augmented Generative AI Chatbots 25

Addressing the Productivity Paradox in Healthcare with Retri...

引用

25th IEEE International Conference on Industrial Technology, ICIT 2024

作者： Ranasinghe, Sajani De Silva, Daswin Mills, Nishan Alahakoon, Damminda Manic, Milos Lim, Yen Ranasinghe, Weranja Centre For Data Analytics And Cognition La Trobe University Melbourne Australia Virginia Commonwealth University Department Of Computer Science Richmond United States Department Of Urology Melbourne Australia

ISBN: (纸本)9798350340266

Artificial Intelligence (AI) is reshaping the health-care landscape through diverse innovations, personalisations and decision-making capabilities. The human-like intelligence of Generative AI has been fundamental in driving this transformation across the sector. Despite large investments and some early successes, several studies have signalled the emergence of a productivity paradox due to inherent limitations of Generative AI that disintegrate within the complexity of healthcare systems and operations. In this study, we investigate the capabilities of Retrieval Augmented Generation (RAG) and Generative AI chatbots in addressing some of these challenges. We present the design and development of a Retrieval Augmented Generative AI Chatbot framework for consultation summaries, diagnostic insights, and emotional assessments of patients. We further demonstrate the technical value of this framework in service innovation, patient engagement and workflow efficiencies that collectively move to address the productivity paradox of AI in healthcare. © 2024 IEEE.

关键词： Decision making

来源：评论

学校读者我要写书评

暂无评论

Forecasting Smart Grid Stability Using Machine Learning Models

Forecasting Smart Grid Stability Using Machine Learning Mode...

引用

2024 IEEE International Conferences of AI, science, engineering, and Technology, AIxSET 2024

作者： Kandukuri, Raja Banad, Yaser M. Southern University Department of Computer Science Baton RougeLA United States School of Electrical and Computer Engineering University of Oklahoma NormanOK United States

ISBN: (纸本)9798350390995

This paper presents a comprehensive study on designing and evaluating machine learning models for forecasting smart power grid stability. The stability of power grids is crucial for balancing electricity supply and demand, but it is challenged by the increasing use of renewable resources and dynamic pricing. Concepts like Decentralized Smart Grid Control (DSGC) aim to address this issue. We developed and compared machine learning models, including Support Vector Machines (SVM), LightGBM, and deep learning architectures, to identify the most accurate approach for forecasting grid stability. Our results showed that the SVM model achieved the highest accuracy of 99.9%, outperforming other models. These outcomes provide valuable insights and practical recommendations for deploying effective forecasting models in real-world smart grid scenarios, ensuring stability, efficient energy management, and seamless integration of renewable resources. © 2024 IEEE.

关键词： Power grids

来源：评论

学校读者我要写书评

暂无评论

Secure and Scalable IoT Networks: Optimizing Blockchain and SDN for Smart Environments 4

Secure and Scalable IoT Networks: Optimizing Blockchain and ...

引用

4th International Conference on Blockchain Technology and Information Security, ICBCTIS 2024

作者： Jain, Arpit Kumar Khatri, Dignesh Ayyagiri, Aravind Mokkapati, Chandrasekhara Bhimanapati, Vijay Bhasker Reddy Alzubaidi, laith H. Koneru Lakshmaiah Education Foundation Department of Computer Science & Engineering Andhra Pradesh India Gujarat University Department of Computer Science & Engineering Gujarat India Wichita State University Department of Computer Science & Engineering DublinCA United States Jawaharlal Nehru Technological University Department of Cs & It Hyderabad India Southern University and A&m College Department of Computer Science LA United States The Islamic University Najaf Iraq Najaf Iraq

ISBN: (纸本)9798331528324

As a result of the Internet of Things, high-speed data transmission and ultra-low latency are achieved in various applications. Data tampering in IoT networks can be caused by malicious or accidental interference, however. By combining blockchain technology with software-defined networking (SDN), we can mitigate these problems. IoT devices and SDN controllers will be able to communicate peer-to-peer by using public and private blockchains. For devices with limited resources, it is an ideal solution since it eliminates Proof-of-Work (PoW) and incorporates distributed trust. In this paper, we present a new Proof-of-Authority (PoA) consensus algorithm that is designed to improve the consistency and reliability of edge devices within the IoT ecosystem, as well as to ensure a high level of trust across IoT ecosystems. During experiments, the integration of SDN and blockchain outperformed existing models in terms of throughput, delay, response time, and CPU utilization, including Blockchain Fundamentals (BCF), AS Cooperative Interdomain Reputation (ASCIR), and Blockchain-based SDN-enabled Secure Routing (BSDNSR). © 2024 IEEE.

关键词： Blockchain

来源：评论

学校读者我要写书评

暂无评论

TSEML: A task-specific embedding-based method for few-shot classification of cancer molecular subtypes

TSEML: A task-specific embedding-based method for few-shot c...

引用

2024 IEEE International Conference on Bioinformatics and Biomedicine, BIBM 2024

作者： Su, Ran Shi, Rui Cui, Hui Xuan, Ping Fang, Chengyan Feng, Xikang Jin, Qiangguo Tianjin University School of Computer Software College of Intelligence and Computing Tianjin China La Trobe University Department of Computer Science and Information Technology Melbourne Australia Shantou University School of Engineering Department of Computer Science Guangdong China Northwestern Polytechnical University School of Software Shaanxi China Yangtze River Delta Research Institute Northwestern Polytechnical University Taicang China

ISBN: (纸本)9798350386226

Molecular subtyping of cancer is recognized as a critical and challenging upstream task for personalized therapy. Existing deep learning methods have achieved significant performance in this domain when abundant data samples are available. However, the acquisition of densely labeled samples for cancer molecular subtypes remains a significant challenge for conventional data-intensive deep learning approaches. In this work, we focus on the few-shot molecular subtype prediction problem in heterogeneous and small cancer datasets, aiming to enhance precise diagnosis and personalized treatment. We first construct a new few-shot dataset for cancer molecular subtype classification and auxiliary cancer classification, named TCGA Few-Shot, from existing publicly available datasets. To effectively leverage the relevant knowledge from both tasks, we introduce a task-specific embedding-based meta-learning framework (TSEML). TSEML leverages the synergistic strengths of a model-agnostic meta-learning (MAML) approach and a prototypical network (ProtoNet) to capture diverse and fine-grained features. Comparative experiments conducted on the TCGA FewShot dataset demonstrate that our TSEML framework achieves superior performance in addressing the problem of few-shot molecular subtype classification. © 2024 IEEE.

关键词： Zero-shot learning

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：