检索结果-内蒙古大学图书馆

Multi-Source Data with Laplacian Eigenmaps and Denoising Autoencoder for Predicting Microbe-Disease Associations via Convolutional Neural Network

引用

Journal of computer science & technology 2025年第2期40卷 588-604页

作者： Xiu-Juan Lei Ya-Li Chen Yi Pan School of Computer Science Shaanxi Normal UniversityXi’an 710119China Faculty of Computer Science and Control Engineering Shenzhen Institute of Advanced TechnologyChinese Academy of SciencesShenzhen 518055China Shenzhen Key Laboratory of Intelligent Bioinformatics Shenzhen Institute of Advanced TechnologyChinese Academy of ScienceShenzhen 518055China

1 School of computer science,Shaanxi Normal University,Xi’an 710119,China 2 Faculty of computer science and Control engineering,Shenzhen Institute of Advanced technology,Chinese academy of sciences,Shenzhen 518055,China 3 Shenzhen Key Laboratory of Intelligent Bioinformatics,Shenzhen Institute of Advanced technology,Chinese academy of science,Shenzhen 518055,China E-mail:xjlei@***;yalichen@***;***@*** Received December 9,2022;accepted July 29,*** Identifying microbes associated with diseases is important for understanding the pathogenesis of diseases as well as for the diagnosis and treatment of *** this article,we propose a method based on a multi-source association network to predict microbe-disease associations,named ***,a heterogeneous network of multimolecule associations is constructed based on associations between microbes,diseases,drugs,and ***,the graph embedding algorithm Laplacian eigenmaps is applied to the association network to learn the behavior features of microbe nodes and disease *** the same time,the denoising autoencoder(DAE)is used to learn the attribute features of microbe nodes and disease ***,attribute features and behavior features are combined to get the final embedding features of microbes and diseases,which are fed into the convolutional neural network(CNN)to predict the microbedisease *** results show that the proposed method is more effective than existing *** addition,case studies on bipolar disorder and schizophrenia demonstrate good predictive performance of the MMHN-MDA model,and further,the results suggest that gut microbes may influence host gene expression or compounds in the nervous system,such as neurotransmitters,or metabolites that alter the blood-brain barrier.

关键词： multi-source data Laplacian eigenmap denoising autoencoder convolutional neural network microbe-disease association

来源：评论

学校读者我要写书评

暂无评论

Neuromorphic-computing-based adaptive learning using ion dynamics in flexible energy storage devices

引用

National science Review 2022年第11期9卷 172-181页

作者： Shufang Zhao Wenhao Ran Zheng Lou Linlin Li Swapnadeep Poddar Lili Wang Zhiyong Fan Guozhen Shen State Key Laboratory for Superlattices and Microstructures Institute of SemiconductorsChinese Academy of Sciences and Center of Materials Science and Optoelectronic EngineeringUniversity of Chinese Academy of Sciences Department of Electronic and Computer Engineering The Hong Kong University of Science and Technology School of Integrated Circuits and Electronics Beijing Institute of Technology

High-accuracy neuromorphic devices with adaptive weight adjustment are crucial for high-performance computing. However, limited studies have been conducted on achieving selective and linear synaptic weight updates without changing electrical pulses. Herein, we propose high-accuracy and self-adaptive artificial synapses based on tunable and flexible MXene energy storage devices. These synapses can be adjusted adaptively depending on the stored weight value to mitigate time and energy loss resulting from recalculation. The resistance can be used to effectively regulate the accumulation and dissipation of ions in single devices, without changing the external pulse stimulation or preprogramming, to ensure selective and linear synaptic weight updates. The feasibility of the proposed neural network based on the synapses of flexible energy devices was investigated through training and machine learning. The results indicated that the device achieved a recognition accuracy of ～95% for various neural network calculation tasks such as numeric classification.

关键词： neuromorphic computing MXene flexible device high accuracy adaptability ion dynamics

来源：评论

学校读者我要写书评

暂无评论

Adaptive energy-preserving algorithms for guiding center system

引用

Plasma science and technology 2023年第4期25卷 11-22页

作者：朱贝贝刘健张嘉炜祝爱卿唐贻发 Department of Applied Mathematics School of Mathematics and PhysicsUniversity of Science and Technology BeijingBeijing 100083Peopleʼs Republic of China School of Nuclear Science and Technology University of Science and Technology of ChinaHefei 230026Peopleʼs Republic of China Advanced algorithm Joint Lab Shandong Computer Science CenterQilu University of TechnologyJinan 250014Peopleʼs Republic of China The State Key Laboratory of Scientific and Engineering Computing The Institute of Computational Mathematics and Scientific/Engineering ComputingAcademy of Mathematics and Systems ScienceChinese Academy of SciencesBeijing 100190Peopleʼs Republic of China School of Mathematical Sciences University of Chinese Academy of SciencesBeijing 100049Peopleʼs Republic of China

We develop two types of adaptive energy preserving algorithms based on the averaged vector field for the guiding center dynamics,which plays a key role in magnetized *** adaptive scheme is applied to the Gauss Legendre’s quadrature rules and time stepsize respectively to overcome the energy drift problem in traditional energy-preserving *** new adaptive algorithms are second order,and their algebraic order is carefully *** results show that the global energy errors are bounded to the machine precision over long time using these adaptive algorithms without massive extra computation cost.

关键词： guiding center system energy-preserving methods adaptive energy conservation

来源：评论

学校读者我要写书评

暂无评论

TCGC: Tweet Classification and Recommendation for Geographical Catastrophes and Events 1

引用

15th International Conference on Soft Computing and Pattern Recognition, SoCPaR 2023 and 14th World Congress on Nature and Biologically Inspired Computing, NaBIC 2023

作者： Aravind Krishnan, A. Deepak, Gerard Santhanavijayan, A. Department of Computer Science and Engineering Manipal Academy of Higher Education Manipal Institute of Technology Bengaluru India Department of Computer Science and Engineering BMS Institute of Technology and Management Karnataka Bengaluru India National Institute of Technology Tiruchirappalli India

ISBN: (数字)9783031810862

ISBN: (纸本)9783031810855

A need for a strategic framework for tweet classification and recommendation has emerged, specifically for geographical catastrophes and events. This paper proposes a semantics-oriented model for disaster tweet classification in which the latent semantic indexing and WikiData are encompassed for topic modelling and auxiliary knowledge addition into the model. Event ontologies highlighting catastrophes also contribute to the information present in the framework. This model encompasses the DT classifier and the transformer classifier in order to classify the dataset and the metadata respectively. The metadata generation also enriches the framework’s knowledge. The framework uses Google’s Knowledge Graph API to generate knowledge graphs and a shuffling agent helps in shuffling the knowledge for diversifying the intermediate solutions. Twitter Semantic Similarity and Adaptive Pointwise Mutual Information (APMI) measures are strong semantics-oriented reasoning models used in the framework for semantic relatedness computation. Animal Migration Optimization which is a metaheuristic optimization framework is also used for the computation of the optimal solution. The paper has a precision of 96.54%, an overall accuracy of 97.02%, and the lowest FDR of 0.04, outperforming all the baseline models considered in this paper. © The Author(s), under exclusive license to Springer Nature Switzerland AG 2025.

关键词： Knowledge graph

来源：评论

学校读者我要写书评

暂无评论

On the Necessity of Collaboration for Online Model Selection with Decentralized Data 38

On the Necessity of Collaboration for Online Model Selection...

引用

38th Conference on Neural Information Processing Systems, NeurIPS 2024

作者： Li, Junfan Wu, Zheshun Xu, Zenglin King, Irwin School of Computer Science and Technology Harbin Institute of Technology Shenzhen China Pengcheng Lab China Institute Fudan University China Shanghai Academy of AI for Science China Department of Computer Science and Engineering The Chinese University of Hong Kong Hong Kong

We consider online model selection with decentralized data over M clients, and study the necessity of collaboration among clients. Previous work proposed various federated algorithms without demonstrating their necessity, while we answer the question from a novel perspective of computational constraints. We prove lower bounds on the regret, and propose a federated algorithm and analyze the upper bound. Our results show (i) collaboration is unnecessary in the absence of computational constraints on clients;(ii) collaboration is necessary if the computational cost on each client is limited to o(K), where K is the number of candidate hypothesis spaces. We clarify the unnecessary nature of collaboration in previous federated algorithms for distributed online multi-kernel learning, and improve the regret bounds at a smaller computational and communication cost. Our algorithm relies on three new techniques including an improved Bernstein's inequality for martingale, a federated online mirror descent framework, and decoupling model selection and prediction, which might be of independent interest. © 2024 Neural information processing systems foundation. All rights reserved.

关键词：

来源：评论

学校读者我要写书评

暂无评论

QTFN:A General End-to-End Time-Frequency Network to Reveal the Time-Varying Signatures of the Time Series

引用

Big Data Mining and Analytics 2024年第3期7卷 905-919页

作者： Tao Chen Yang Jiao Lei Xie Hongye Su State Key Laboratory of Industrial Control Technology Zhejiang UniversityHangzhou 310027China Shenzhen Key Laboratory of Intelligent Bioinformatics and College of Computer Science and Control EngineeringShenzhen Institute of Advanced TechnologyChinese Academy of SciencesShenzhen 518055China

Nonstationary time series are ubiquitous in almost all natural and engineering *** the time-varying signatures from nonstationary time series is still a challenging problem for data *** Time-Frequency Distribution(TFD)provides a powerful tool to analyze these ***,they suffer from Cross-Term(CT)issues that impair the readability of ***,to achieve high-resolution and CT-free TFDs,an end-to-end architecture termed Quadratic TF-Net(QTFN)is proposed in this *** by classic TFD theory,the design of this deep learning architecture is heuristic,which firstly generates various basis functions through ***,more comprehensive TF features can be extracted by these basis ***,to balance the results of various basis functions adaptively,the Efficient Channel Attention(ECA)block is also embedded into ***,a new structure called Muti-scale Residual Encoder-Decoder(MRED)is also proposed to improve the learning ability of the model by highly integrating the multi-scale learning and encoder-decoder ***,although the model is only trained by synthetic signals,both synthetic and real-world signals are tested to validate the generalization capability and superiority of the proposed QTFN.

关键词： Time-Frequency Analysis(TFA) Multi-scale Residual Encoder-Decoder(MRED) quadratic Time-Frequency Distribution(TFD)

来源：评论

学校读者我要写书评

暂无评论

Image Caption And Speech Generation Using LSTM And GTTS API 2

Image Caption And Speech Generation Using LSTM And GTTS API

引用

2nd International Conference on Augmented Intelligence and Sustainable Systems, ICAISS 2023

作者： Ram, K. Bhargav Venkatesh, B. Sree, Sala Pooja Sai Anilkumar, Chunduru Reddy, V. Surya Narayana Kodumuri, Bhavya Bvrit Hyderabad College of Engineering for Women Department of Computer Science and Engineering Hyderabad India Systems Engineer Infosys Hyderabad India Gmr Institute of Technology Dept of Information Technology Andhra Pradesh Rajam532127 India Vnr Vignana Jyothi Institute of Engineering and Technology Department of Computer Science and Engineering Hyderabad India Vidya Jyothi Institute of Technology Department of Computer Science and Engineering HYderabad India

ISBN: (纸本)9798350325799

Image Caption generation is one of the challenging tasks in the field of artificial intelligence. It is used to generate a textual description for a given picture. But due to, the recent advancement in deep learning techniques requires only one single end-to-end model to create a caption for the given image. This paper is intended to provide an image caption and speech generator used to generate a single-line description for a given image and audio/speech for the report generated. The researchers used a sophisticated pipeline of specifically designed models in the previous models. To achieve the proposed model, VGG16 and LSTM models are used to obtain descriptions for the image and GTTS API is used for the audio/speech generation. © 2023 IEEE.

关键词： Long short-term memory

来源：评论

学校读者我要写书评

暂无评论

Optimizing Diagnostic Accuracy in Ultrasound Imaging Through Cloud-Enabled Deep Learning Techniques 4

Optimizing Diagnostic Accuracy in Ultrasound Imaging Through...

引用

4th International Conference on Soft Computing for Security Applications, ICSCSA 2024

作者： Kalpana, V. Girija, M. Tharageswari, K. Yobu, D. Revathi, J. Aruna, R. Department of Computer Science and Engineering Vel Tech Rangarajan Dr.Sagunthala R&D Institute of Science and Technology Chennai India Jeppiaar University School of Engineering and Technology Chennai India Department of Computer Science and Engineering Karpagam Academy of Higher Education Coimbatore India Computer Science and Engineering Cyber Security J.J College of Engineering and Technology Tricy India Department of Mathematics Kongu Engineering College Erode India Department of Computer Science and Engineering School of Computing Vel Tech Rangarajan Dr.Sagunthala R&D Institute of Science and Technology Chennai India

ISBN: (纸本)9798331515720

This research aims to study predicting and optimizing the diagnostic accuracy by cloud-enabled deep learning techniques in ultrasound imaging for kidney disease diagnosis. The main findings are that CNN gives 92% accurate outcomes in terms of sensitivity, precision, specificity, AUC-ROC, and other terms used for predictive outcomes in the current research. The other outcomes are 88% accuracy and 91% specificity in predicting kidney diseases by using the RNN model. The other two models used for learning are transfer learning and federated learning which also provide good outcomes which are 94% accurate and 91% accurately in diagnosing with the above-said models. These learning approaches optimize or increase the accuracy rates such as 93% by the transfer learning approach and 90% accuracy outcomes by the federated learning mechanism. These above-given models have been taken based on advanced technologies that are used to predict or forecast the diseases through learning processes. The success of these learning mechanisms is explained by the reliable processes used in the learning of the models that are possible by the neural network used in the learning mechanisms. The high accuracy rates of the models show the successfulness of such clinics, who are in rural areas and are unable to meet their necessary outcome-oriented goals, then measures can be taken for running tele-clinics using the above models of learning techniques to diagnose the diseases at the correct time by 'n' number of people in rural areas. Mine is one of them working as a health worker for which measures need to be taken for finding an accurate solution that operates on cloud-enabled deep learning. This model help run family life and be able to diagnose certain conditions at desired rural clinics out of the above learning models. These tools of learning are studied on cloud-based deep learning techniques and performed well in diagnosing kidney diseases on rural tele-medicine and these learning techni

关键词： Ultrasonic imaging

来源：评论

学校读者我要写书评

暂无评论

A Paradigm Shift in Real-Time Detection: The YOLOv7 Approach to Trainable Freebie Optimization 3

A Paradigm Shift in Real-Time Detection: The YOLOv7 Approach...

引用

3rd International Conference on Electronics and Renewable Systems, ICEARS 2025

作者： Vinitha, A. Jeevitha, S. Kumar, N.B. Mahesh Sathish Kumar, L. Saravanan, D. Karpagam Academy of Higher Education Department of Computer Applications Tamilnadu Coimbatore India Kalasalingam Academy of Research and Education Department of Computer Science and Information Technology Krishnankoil Tamilnadu Virudhunagar India Hindusthan Institute of Technology Computer Science and Engineering Tamil Nadu Coimbatore India Madhya Pradesh Sehore India

ISBN: (纸本)9798331509675

Significant progress has been made in real-time object detection, with YOLOv7 emerging as a state-of-the-art model that achieves the highest levels of accuracy and speed. This study combines Bag-of-Freebies (BoF) strategies with YOLOv7 to improve performance without incurring extra computational expenses. Our tests show that adding BoF preserves real-time inference capabilities while increasing mAP@0.5 by 3.5%. The accuracy and FPS of YOLOv7 with BoF are superior to those of the current YOLO-based models (YOLOv4, YOLOv5, YOLOv6). According to the results, YOLOv7 with BoF sets a new standard for real-time object detection by achieving a mAP@0.5 of 55.3% and 83 FPS on the COCO dataset. © 2025 IEEE.

关键词： Contrastive Learning

来源：评论

学校读者我要写书评

暂无评论

Gesture Recognition Machine Vision Video Calling Application Using YOLOv8 22

Gesture Recognition Machine Vision Video Calling Application...

引用

22nd International Symposium on Communications and Information Technologies, ISCIT 2023

作者： Pruthvi Darshan, S.S. Patil, Shantakumar B. Patil, Bhargav S. Premjyoti Sai Vidya Institute of Technology Department of Computer Science & Engineering Karnataka Bengaluru560064 India B.M.S. College of Engineering Department of Computer Science & Engineering Karnataka Bengaluru560019 India Karnataka Bengaluru560019 India

ISBN: (纸本)9781665457316

Gesture recognition pertains to a computer system's capacity to identify and understand hand gestures, categorizing them into captions or text. This technology holds vast potential for applications in education, communication, and accessibility, especially for individuals with hearing disabilities. This research introduces a gesture recognition video calling application, enabling communication among individuals with and without speech and hearing abilities. To enhance accuracy, the application employs Darknet's YOLOv8 for object detection. YOLOv8 is a robust and efficient system that excels in various computer vision tasks, such as object detection, instance segmentation, and image classification. The paper proposes a trained model for image classification, encompassing ten different hand gesture classes, which processes each video frame during the call, achieving a remarkable 98% accuracy with a low latency of 10ms. The application displays the classification results on top of the interface during the video call. © 2023 IEEE.

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：