检索结果-内蒙古大学图书馆

An information-volume-based distance measure for decision-making

Chinese Journal of Aeronautics 2023年第5期36卷 392-405页

作者： Zhanhao ZHANG Fuyuan XIAO School of Big Data and Software Engineering Chongqing UniversityChongqing 401331China

D-S evidence theory,as a general framework for reasoning with uncertainty,allows combining pieces of evidence from different information sources to derive a degree of belief function that is a type of fuzzy ***,the mass assignments given by unknown information sources are *** to measure the difference between the mass assignments has aroused people’s *** this paper,inspired by the information volume,a novel distance-based measure is proposed to measure the difference between mass *** method can refine the uncertain information given by experts and compare the refined information to obtain the difference between mass *** the same time,it is verified that the measure not only meets the properties of distance,but also proves the superiority of the proposed Information Volume Distance(IVD)through simulation ***,in the process of information fusion,the reliability of each source could be quantified through ***,based on IVD,a new multi-source information algorithm is proposed to solve the problem of multi-source information ***,algorithm is applied to decision-making problem and compare with other methods to verify the effectiveness.

关键词： Basic belief assignments Decision-making Distance measure Evidence theory Multi-source information fusion

来源：评论

学校读者我要写书评

暂无评论

IoTDQ: An Industrial IoT data Analysis Library for Apache IoTDB

引用

big data Mining and Analytics 2024年第1期7卷 29-41页

作者： Pengyu Chen Wendi He Wenxuan Ma Xiangdong Huang Chen Wang School of Software Tsinghua UniversityBeijing 100084China National Engineering Research Center for Big Data Software(NERCBDS) Tsinghua UniversityBeijing 100084China

There is a growing demand for time series data analysis in industry *** loTDB is a time series database designed for the Internet of Things(loT)with enhanced storage and I/O *** User-Defined Functions(UDF)provided,computation for time series can be executed on Apache loTDB *** satisfy most of the common requirements in industrial time series analysis,we create a UDF library,loTDQ,on Apache *** library integrates stream computation functions on data quality analysis,data profiling,anomaly detection,data repairing,*** enables users to conduct a wide range of analyses,such as monitoring,error diagnosis,equipment reliability *** provides a framework for users to examine loT time series with data quality *** show that loTDQ keeps the same level of performance compared to mainstream alternatives,and shortens I/O consumption for Apache loTDB users.

关键词： industrial big data data quality data mining and analytics

来源：评论

学校读者我要写书评

暂无评论

FedDAGC: Dynamic Adaptive Graph Coarsening for Federated Learning on Non-IID Graphs 18th

FedDAGC: Dynamic Adaptive Graph Coarsening for Federated Le...

引用

18th International Conference on Wireless Artificial Intelligent Computing Systems and Applications, WASA 2024

作者： Zhang, Chengxi Hu, Chunqiang College of Big Data and Software Engineering Chongqing University Chongqing China

ISBN: (纸本)9783031714665

Federated graph learning (FGL) has demonstrated great potential in collaboratively training graph neural network (GNN) models under the federated learning (FL) framework, benefiting from its advantage of privacy preservation. However, FGL suffers from the significant challenge of non-IID problem. In the real world, cross-source or even cross-domain graph data frequently exist among clients. To tackle this challenge, we propose a novel FGL framework, FedDAGC, from the new perspective of graph coarsening and consensus. Specifically, our method inserts a dynamic adaptive graph coarsening (DAGC) module into local GNN models to generate coarsened graphs as augmented inputs for subsequent model training. By sharing the intermediate parameters in DAGC instead of traditional GNN model parameters, FedDAGC enables clients to reach a consensus on coarsening directions, for extracting crucial patterns from local graphs based on common core knowledge while protecting privacy. Extensive experiments and analyses on various non-IID settings demonstrate the efficacy and superiority of FedDAGC. © The Author(s), under exclusive license to Springer Nature Switzerland AG 2025.

关键词： Federated learning

来源：评论

学校读者我要写书评

暂无评论

Multivariate Time Series Forecasting with Transfer Entropy Graph

引用

Tsinghua Science and Technology 2023年第1期28卷 141-149页

作者： Ziheng Duan Haoyan Xu Yida Huang Jie Feng Yueyang Wang School of Big Data and Software Engineering Chongqing UniversityChongqing 401331China College of Control Science and Engineering Zhejiang UniversityHangzhou 310027China

Multivariate Time Series(MTS)forecasting is an essential problem in many *** forecasting results can effectively help in making *** date,many MTS forecasting methods have been proposed and widely ***,these methods assume that the predicted value of a single variable is affected by all other variables,ignoring the causal relationship among *** address the above issue,we propose a novel end-to-end deep learning model,termed graph neural network with neural Granger causality,namely CauGNN,in this *** characterize the causal information among variables,we introduce the neural Granger causality graph in our *** variable is regarded as a graph node,and each edge represents the casual relationship between *** addition,convolutional neural network filters with different perception scales are used for time series feature extraction,to generate the feature of each ***,the graph neural network is adopted to tackle the forecasting problem of the graph structure generated by the *** benchmark datasets from the real world are used to evaluate the proposed CauGNN,and comprehensive experiments show that the proposed method achieves state-of-the-art results in the MTS forecasting task.

关键词： Multivariate Time Series(MTS)forecasting neural Granger causality graph Transfer Entropy(TE)

来源：评论

学校读者我要写书评

暂无评论

Time-domain distributed multichannel wiener filtering speech enhancement algorithm

Time-domain distributed multichannel wiener filtering speech...

引用

2024 International Workshop on Automation, Control, and Communication engineering, IWACCE 2024

作者： Tu, Jingxian Li, Lianfen Lv, Haifeng School of Big Data and Software Engineering Wuzhou University Wuzhou China

ISBN: (数字)9781510685420

ISBN: (纸本)9781510685413

To recover pure speech from observed speech, this paper presents a time-domain multi-channel Wiener filter speech enhancement algorithm for the distributed speech model. For reducing the noise from noisy speech in time domain, this paper first gives the formula about the energy of noise reduction and speech distortion, then establishes the optimization problem on noise reduction and speech distortion, and finally solves the optimization problem to get the formula of the optimal linear filter. In addition, this paper uses an iterative method to estimate the source speech signal autocorrelation matrix, which improves the estimation accuracy. Simulation experiment results show the proposed algorithm can obtain the better performance than several classical multi-channel speech enhancement algorithms. © 2024 SPIE. All rights reserved.

关键词： Wiener filtering

来源：评论

学校读者我要写书评

暂无评论

Progressively Relaxed Knowledge Distillation

Progressively Relaxed Knowledge Distillation

引用

2024 International Joint Conference on Neural Networks, IJCNN 2024

作者： Gu, Xiaozhe Jin, Ran Li, Mingliang Zhejiang Wanli University College of Big Data and Software Engineering China

ISBN: (纸本)9798350359312

Knowledge Distillation (KD) enhances the generalization ability of a student model by transferring knowledge from a teacher model. However, literature suggests that the student may struggle to match the teacher's predictions perfectly, even with sufficient capacity. A distillation scheme imposing strong constraints on the student may lead to over-regularization and performance degradation. On the other hand, a scheme with low supervision intensity can alleviate over-regularization but may lead to the student acquiring less potentially beneficial knowledge. Determining the optimal distillation scheme often requires significant effort and experimentation. Furthermore, the fact that the optimal distillation strategy may vary at different stages of the distillation process is also often *** this work, we introduce PRKD (Progressively Relaxed Knowledge Distillation), an advanced distillation framework that dynamically adjusts the supervision intensity from the teacher throughout the training process. Specifically, PRKD is built upon a gradual transition of the distillation scheme, progressing iteratively from a strong to a weak configuration. Through this transition, the student preserves essential knowledge from the teacher while simultaneously alleviating the potential for excessive regularization. Our comprehensive experiments on diverse datasets demonstrate the efficacy of PRKD, consistently surpassing other state-of-the-art methods. © 2024 IEEE.

关键词： Students

来源：评论

学校读者我要写书评

暂无评论

Research on the Relationship Between New Energy Vehicles and Dual Carbon in Yangtze River Delta Based on GRA-LSTM Model 7

Research on the Relationship Between New Energy Vehicles and...

引用

7th IEEE Advanced Information Technology, Electronic and Automation Control Conference, IAEAC 2024

作者： Zhou, Jinxuan Deng, Liang Tong, Yi Gao, Yan Wuhan Business University Department of Information Engineering Hubei China Data Science and Big Data Technology China Internet of Things Engineering China Software Engineering Big Data Technology

ISBN: (纸本)9798350339161

New energy automobile industry plays an important role in building a green, low-carbon and recycling industrial system. In this paper, the prediction simulation training and prediction accuracy comparison study are carried out with the help of the newly constructed GRA-LSTM model, Biological Neural Network model and the first-order one-variable gray GM (1, 1) model. The LSTM model is created and trained by Learning Rate Decay function. The Learning Rate Decay callback function is set, and the learning rate is gradually reduced during the training process to carry out simulation training, and finally the trained model is used to make predictions, and the time for the Yangtze River Delta region to reach carbon peak is 2028, and the time to reach carbon neutrality is 2060, and at the same time this paper analyzes and finds out that the higher the new energy automobile market ownership is, the shorter the time for carbon peak and carbon neutrality will be.2. © 2024 IEEE.

关键词： Forecasting

来源：评论

学校读者我要写书评

暂无评论

Optimization of Container Cloud Elastic Scaling Integrating ARIMA Forecasting and HPA Response 4

Optimization of Container Cloud Elastic Scaling Integrating ...

引用

4th International Conference on Communication Technology and Information Technology, ICCTIT 2024

作者： Zhang, Liangbin Wang, Feng Wang, Yuanming Liu, Xinwei Zhejiang Wanli University College of Big Data and Software Engineering Ningbo China

ISBN: (纸本)9798331528973

Load changes of cloud computing resources show more and more complex characteristics. Load forecasting is crucial for elastic scaling of cloud computing resources. Kubernetes is a mainstream container cloud platform. Its built-in HPA policy automatically adjusts the number of Pod replicas according to load changes to achieve elastic scaling, but it has the problems of single load metric and user response delay during peak periods. This paper proposes a container cloud elastic scaling optimization method based on load forecasting and response, which dynamically monitors and calculates the combined load of CPU and Memory in the Pod set, uses ARIMA model to complete the combined load forecasting value based on historical data, and HPA responsive algorithm to make up for the lack of forecasting. Experimental results show that compared with the HPA strategy, our method can execute the corresponding scaling operations in advance, and can effectively reduce the response delay time and improve the Qos quality. © 2024 IEEE.

关键词： Cloud platforms

来源：评论

学校读者我要写书评

暂无评论

Enhancing Cold Chain Logistics Efficiency: Improved GA for Time-Window-Constrained Routing 3

Enhancing Cold Chain Logistics Efficiency: Improved GA for T...

引用

3rd International Conference on Computer Applications Technology, CCAT 2024

作者： Gao, Xubo Wang, Yuanming Zhang, Liangbin Liu, Xinwei Zhejiang wanli University College of Big Data and Software Engineering Ningbo China

ISBN: (纸本)9798331539795

The routing optimization of distribution vehicles, reducing distribution costs and improving customer satisfaction with delivery time are the keys of cold chain logistics. Based on the original example, we propose coexisting soft and hard time windows that are more in line with user needs, and establish a cold chain distribution mathematical model with the minimum total cost under the constraint of vehicle load. SUS, PMX and reinsertion operator are used to improve the genetic algorithm to solve the model. The example experiments show that the proposed method is obviously superior to the traditional method in the total vehicle mileage, the total distribution cost and the solution time. Our strategy improves the problem of premature convergence and local optimum, and has a good effect on solving the problem of cold chain logistics distribution. © 2024 IEEE.

关键词： Customer satisfaction

来源：评论

学校读者我要写书评

暂无评论

MEAHO: Membrane Evolutionary Algorithm for Hyperparameter Optimization of Deep Convolutional Neural Networks 38th

MEAHO: Membrane Evolutionary Algorithm for Hyperparameter Op...

引用

38th CCF National Conference of Computer Applications, CCF NCCA 2023

作者： Mu, Jie Hou, Jiaxin Wan, Ying School of Big Data and Software Engineering Chongqing University Chongqing400044 China

ISBN: (纸本)9789819987634

Membrane algorithm has been used to solve many optimization problems since it was put forward. These methods used the membrane algorithm as a container for other algorithms to solve many problems, such as, traveling salesman problem, the knapsack problem and so on. In this paper, from the angle of membrane algorithm, the solution of hyperparameter optimization problem by membrane algorithm is explored, with the hyperparameter space divided by lattice method, which transformed into membrane structure subsequently. For the evolution of membrane, the entropy reduction principle is proposed. Compared with other membrane algorithms, this paper solves the optimization problem from the membrane algorithm itself, designs the corresponding algorithm for the membrane structure, and obtains the experimental results through CNN experiment. © 2024, The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd.

关键词： Evolutionary algorithms

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：