检索结果-内蒙古大学图书馆

AFSTGCN:Prediction for multivariate time series using an adaptive fused spatial-temporal graph convolutional network

引用

Digital Communications and Networks 2024年第2期10卷 292-303页

作者： Yuteng Xiao Kaijian Xia Hongsheng Yin Yu-Dong Zhang Zhenjiang Qian Zhaoyang Liu Yuehan Liang Xiaodan Li School of Information and Control Engineering China University of Mining and TechnologyXuzhou221116China Department of Informatics University of LeicesterLeicesterLE17RHUK The Affiliated Changshu Hospital of Soochow University Changshu215500China School of Computer Science and Engineering Changshu Institute of TechnologySuzhou215500China

The prediction for Multivariate Time Series(MTS)explores the interrelationships among variables at historical moments,extracts their relevant characteristics,and is widely used in finance,weather,complex industries and other ***,it is important to construct a digital twin ***,existing methods do not take full advantage of the potential properties of variables,which results in poor predicted *** this paper,we propose the Adaptive Fused Spatial-Temporal Graph Convolutional Network(AFSTGCN).First,to address the problem of the unknown spatial-temporal structure,we construct the Adaptive Fused Spatial-Temporal Graph(AFSTG)***,we fuse the spatial-temporal graph based on the interrelationship of spatial ***,we construct the adaptive adjacency matrix of the spatial-temporal graph using node embedding ***,to overcome the insufficient extraction of disordered correlation features,we construct the Adaptive Fused Spatial-Temporal Graph Convolutional(AFSTGC)*** module forces the reordering of disordered temporal,spatial and spatial-temporal dependencies into rule-like *** dynamically and synchronously acquires potential temporal,spatial and spatial-temporal correlations,thereby fully extracting rich hierarchical feature information to enhance the predicted *** on different types of MTS datasets demonstrate that the model achieves state-of-the-art single-step and multi-step performance compared with eight other deep learning models.

关键词： Adaptive adjacency matrix Digital twin Graph convolutional network Multivariate time series prediction Spatial-temporal graph

来源：评论

学校读者我要写书评

暂无评论

Vpit: real-time embedded single object 3D tracking using voxel pseudo images

引用

Neural Computing and Applications 2024年第32期36卷 20341-20354页

作者： Oleksiienko, Illia Nousi, Paraskevi Passalis, Nikolaos Tefas, Anastasios Iosifidis, Alexandros Department of Electrical and Computer Engineering Aarhus University Finlandsgade Aarhus8200 Denmark Department of Informatics Aristotle University of Thessaloniki University Campus Thessaloniki54124 Greece Department of Chemical Engineering Aristotle University of Thessaloniki University Campus Thessaloniki54124 Greece

In this paper, we propose a novel voxel-based 3D single object tracking (3D SOT) method called Voxel Pseudo Image Tracking (VPIT). VPIT is the first method that uses voxel pseudo images for 3D SOT. The input point cloud is structured by pillar-based voxelization, and the resulting pseudo image is used as an input to a 2D-like Siamese SOT method. The pseudo image is created in the Bird’s-eye View (BEV) coordinates;and therefore, the objects in it have constant size. Thus, only the object rotation can change in the new coordinate system and not the object scale. For this reason, we replace multi-scale search with a multi-rotation search, where differently rotated search regions are compared against a single target representation to predict both position and rotation of the object. Experiments on KITTI [1] Tracking dataset show that VPIT is the fastest 3D SOT method and maintains competitive Success and Precision values. Application of a SOT method in a real-world scenario meets with limitations such as lower computational capabilities of embedded devices and a latency-unforgiving environment, where the method is forced to skip certain data frames if the inference speed is not high enough. We implement a real-time evaluation protocol and show that other methods lose most of their performance on embedded devices;while, VPIT maintains its ability to track the object. © The Author(s) 2024.

关键词： 3D tracking Embedded deep learning (DL) Pillars Pseudo images Real-time neural networks Single object tracking Voxels

来源：评论

学校读者我要写书评

暂无评论

Interactive Learning Through Virtual Reality Operating Training Simulations for Industry Operators 1

Interactive Learning Through Virtual Reality Operating Train...

引用

1st IEEE International Symposium on Consumer Technology, ISCT 2024

作者： Hanifati, Kirana Achmad, Zacky Maulana Arissabarno, Cahyo Fikri, Muhammad Ali Razandistiawan, Adzhani Muhammad, Farhan Politeknik Elektronika Negeri Surabaya Department of Informatics and Computer Engineering Surabaya Indonesia Pt Molca Teknologi Nusantara Surabaya Indonesia

ISBN: (纸本)9798350365191

In the manufacturing industry, factory operations encounter three primary conditions: normal, start-up, and shutdown, each necessitating specialized handling for efficiency and safety. Manual training to address these conditions often proves suboptimal, intricate, time-consuming, and poses high safety risks. To tackle this issue, this research proposes and develops a Virtual Reality (VR)-based Operator Training Simulator (OTS). The VR OTS enables interactive and realistic simulation training from the perspective of both field operators and control panel operators. he development process involved iterative design and testing phases, incorporating feedback from a sample of 50 operators to ensure usability and realism. Data analysis was performed to evaluate the simulator's impact on training efficiency and effectiveness. Research findings indicate that this VR application successfully generates highly realistic simulations for normal, start-up, and shutdown conditions, enhancing operator readiness and skills. Its performance is also adequate with a frame rate of 72 frames per second (FPS) and a manageable CPU and GPU load. Further studies are warranted to optimize OTS with other immersive technologies to bolster training efficiency. © 2024 IEEE.

关键词： Virtual environments

来源：评论

学校读者我要写书评

暂无评论

Comparative Analysis of Memory-Based Collaborative Filtering and Deep Learning Models for Resolving Cold Start and Data Sparsity Issues in E-commerce Recommender Systems 3

Comparative Analysis of Memory-Based Collaborative Filtering...

引用

3rd IEEE World Conference on Applied Intelligence and Computing, AIC 2024

作者： Shetty, Amrithkala M Manjaiah, D.H. Aljunid, Mohammed Fadhel Yogesh, K.M. Mangalore University Department of Computer Science Karnataka India Thamar University Computer and Informatics Center Thamar Yemen M S Ramaiah University of Applied Sciences Department of Computer Science and Engineering India

ISBN: (纸本)9798350384598

In today's digital landscape, where online activities and e-commerce play pivotal roles, Recommender Systems (RS) have become indispensable tools for enhancing user experiences by providing personalized recommendations. However, collaborative filtering (CF) algorithms often face challenges such as the cold start problem and data sparsity. This work addresses these challenges using memory-based CF approaches such as Singular Value Decomposition (SVD), SVD++, K-Nearest Neighbors (KNN) Baseline, and K-Nearest Neighbors (KNN) Basic. Hyperparameters for these algorithms are optimized using GridSearchCV with cross-validation values set to 5, 10, and 15. Additionally, Deep Learning (DL) models including LSTM, BiLSTM, Att+BiLSTM, User-Product Fusion (UPF), and Deep Collaborative Filtering (DCF) are explored. The BiLSTM model achieves a lower Root Mean Squared Error (RMSE) of 1.1224, while the DCF model achieves a Mean Absolute Error (MAE) of 0.6897. Among the evaluated models, SVD++ achieves an RMSE of 0.9881 and an MAE of 0.7003 with cross-validation set to 15, indicating superior predictive accuracy. The best performance is observed with cross-validation set to 15. These results highlight the effectiveness of optimized memory-based CF algorithms in overcoming the cold start and data sparsity issues, leading to more accurate and reliable recommendations. The findings suggest that memory-based CF algorithms, specifically SVD++, can significantly enhance the development of robust recommender systems, ultimately improving user experience in e-commerce and other online platforms. © 2024 IEEE.

关键词： Collaborative filtering

来源：评论

学校读者我要写书评

暂无评论

Optimizing Item-based Collaborative Filtering for Course Recommendation: A Case Study with Coursera Data

Optimizing Item-based Collaborative Filtering for Course Rec...

引用

2024 International Conference on Information Technology Systems and Innovation, ICITSI 2024

作者： Suryani, Mira Pangestu, Risyad Pradana, Aditya Universitas Padjadjaran Departement of Computer Science Sumedang Indonesia Institut Teknologi Bandung School of Electrical Engineering and Informatics Bandung Indonesia

ISBN: (纸本)9798331511470

In e-learning environments, the sheer volume of available courses often overwhelms users, leading to decision fatigue and disengagement Learners vary widely in their preferences, learning styles, and goals, yet generic recommendations fail to account for these differences, resulting in suboptimal learning experiences. Therefore, personalized recommendations which are generally found in recommendation systems are needed especially in e-learning systems. Collaborative filtering is a popular method in recommendation systems, generating predictions by analyzing user ratings to forecast items of interest. This study implements an item-based collaborative filtering approach, calculating item similarity to recommend courses effectively. The study involves developing a web-based recommendation system utilizing a dataset of 623 Coursera courses and 1,454,711 course review entries. Prediction accuracy is enhanced through weighted sum and adjusted weighted sum calculations, employing cosine similarity and adjusted cosine similarity to refine similarity scores. Prediction errors are evaluated using Mean Absolute Error (MAE) and Root Mean Square Error (RMSE) metrics. Testing various K-nearest neighbor values reveals that an optimal K value of 150 yields an MAE of 0.4791 and an RMSE of 0.7882 for weighted sum predictions. For adjusted weighted sum predictions, an optimal K value of 160 produces an MAE of 0.4792 and an RMSE of 0.7638. These results highlight the effectiveness of item-based collaborative filtering in delivering accurate course recommendations and showcase the impact of similarity measures on predictive accuracy. © 2024 IEEE.

关键词： Collaborative filtering

来源：评论

学校读者我要写书评

暂无评论

Optimization of Cooperative RelayingMolecular Communications for Nanomedical Applications

引用

computer Modeling in engineering & Sciences 2024年第2期138卷 1259-1275页

作者： Eman S.Attia Ashraf A.M.Khalaf Fathi E.Abd El-Samie Saied M.Abd El-atty Konstantinos A.Lizos Osama Alfarraj Heba M.El-Hoseny Department of Electrical Engineering Faculty of EngineeringMinia UniversityMinia61111Egypt Department of Electronics and Electrical Communications Engineering Faculty of Electronic EngineeringMenoufia UniversityMenouf32952Egypt Department of Informatics Faculty of Mathematics and Natural SciencesUniversity of Oslo(UiO)OsloNorway Computer Science Department Community CollegeKing Saud UniversityRiyadh11437Saudi Arabia Department of Computer science The Higher Future Institute for Specialized Technological StudiesEl ShoroukEgypt

Recently,nano-systems based on molecular communications via diffusion(MCvD)have been implemented in a variety of nanomedical applications,most notably in targeted drug delivery system(TDDS)***,because the MCvD is unreliable and there exists molecular noise and inter symbol interference(ISI),cooperative nano-relays can acquire the reliability for drug delivery to targeted diseased cells,especially if the separation distance between the nano transmitter and nano receiver is *** this work,we propose an approach for optimizing the performance of the nano system using cooperative molecular communications with a nano relay scheme,while accounting for blood flow effects in terms of drift *** fractions of the molecular drug that should be allocated to the nano transmitter and nano relay positioning are computed using a collaborative optimization problem solved by theModified Central Force Optimization(MCFO)*** the previous work,the probability of bit error is expressed in a closed-form *** is used as an objective function to determine the optimal velocity of the drug molecules and the detection threshold at the nano *** simulation results show that the probability of bit error can be dramatically reduced by optimizing the drift velocity,detection threshold,location of the nano-relay in the proposed nano system,and molecular drug budget.

关键词： Nanomedical system molecular communication cooperative relay optimization

来源：评论

学校读者我要写书评

暂无评论

Traffic Control Based on Integrated Kalman Filtering and Adaptive Quantized Q-Learning Framework for Internet of Vehicles

引用

computer Modeling in engineering & Sciences 2024年第3期138卷 2103-2127页

作者： Othman S.Al-Heety Zahriladha Zakaria Ahmed Abu-Khadrah Mahamod Ismail Sarmad Nozad Mahmood Mohammed Mudhafar Shakir Hussein Alsariera Faculty of Electronic and Computer Engineering Universiti Teknikal Malaysia Melaka(UTeM)Durian TunggalMalaysia College of Computing&Informatics Saudi Electronic UniversityRiyadhSaudi Arabia Department of Electrical Electronic and System EngineeringUniversiti Kebangsaan MalaysiaUKM BangiSelangorMalaysia Computer Technology Engineering College of Engineering TechnologyAl-Kitab UniversityKirkukIraq Faculty of Engineering and Electronic Engineering Universiti Tun Hussein Onn Malaysia(UTHM)Batu PahatMalaysia 6Research Center DepartmentThe University of MashreqBaghdadIraq

Intelligent traffic control requires accurate estimation of the road states and incorporation of adaptive or dynamically adjusted intelligent algorithms for making the *** this article,these issues are handled by proposing a novel framework for traffic control using vehicular communications and Internet of Things *** framework integrates Kalman filtering and *** smoothing Kalman filtering,our data fusion Kalman filter incorporates a process-aware model which makes it superior in terms of the prediction *** traditional Q-learning,our Q-learning algorithm enables adaptive state quantization by changing the threshold of separating low traffic from high traffic on the road according to the maximum number of vehicles in the junction *** evaluation,the model has been simulated on a single intersection consisting of four roads:east,west,north,and south.A comparison of the developed adaptive quantized Q-learning(AQQL)framework with state-of-the-art and greedy approaches shows the superiority of AQQL with an improvement percentage in terms of the released number of vehicles of AQQL is 5%over the greedy approach and 340%over the state-of-the-art ***,AQQL provides an effective traffic control that can be applied in today’s intelligent traffic system.

关键词： Q-learning intelligent transportation system(ITS) traffic control vehicular communication kalman filtering smart city Internet of Things

来源：评论

学校读者我要写书评

暂无评论

Dynamic Client Selection for Over-the-Air Federated Learning Network

引用

IEEE Internet of Things Journal 2025年第12期12卷 18508-18519页

作者： Shi, Fang Lin, Weiwei Peng, Chaoda Zhong, Cankun Wang, Dong Cheng, Mingyue South China Agricultural University College of Mathematics and Informatics Guangzhou510006 China South China University of Technology School of Computer Science and Engineering Guangzhou510006 China Peng Cheng Laboratory Shenzhen518066 China

As a privacy-preserving solution, federated learning (FL) demonstrates great potential in distributed model training, but limited bandwidth, particularly in near-field communication (NFC)-based systems, emerges as a key bottleneck by restricting the number of participating clients. To address this challenge, over-the-air FL leverages the superposition property of wireless multiple-access channels, enabling faster model training and accommodating more clients, even in bandwidth-constrained scenarios like NFC. However, due to its analog-integrated nature, the FL performance is also affected by other factors, such as channel noise. These motivate us to consider how the selected client set and channel noise affect FL performance. To explore this concern, in this article, we consider an over-the-air FL system with analog gradient aggregation and analyze the impact of the selected client set and channel noise on FL training performance. The theoretical analysis effectively shows the importance of the clients’ number and the power scaling factor to the FL training performance. Based on the theoretical analysis, we transform the global optimization problem into the client selection problem and propose a dynamic client selection scheme to optimize the training performance under the aggregation error constraint. Experimental results demonstrate that our proposed scheme can boost FL by speeding up the convergence of the global model (at least 35%) and saving energy consumption. © 2014 IEEE.

关键词： Mean square error

来源：评论

学校读者我要写书评

暂无评论

A Deep Learning Approach to Mesh Segmentation

引用

computer Modeling in engineering & Sciences 2023年第5期135卷 1745-1763页

作者： Abubakar Sulaiman Gezawa Qicong Wang Haruna Chiroma Yunqi Lei Department of Computer Science School of InformaticsXiamen UniversityXiamen361005China Shenzen Research Institute Xiamen UniversityShenzen518000China College of Computer Science and Engineering University of Hafr Al BatinHafr Albatin39524Saudi Arabia

In the shape analysis community,decomposing a 3D shape intomeaningful parts has become a topic of interest.3D model segmentation is largely used in tasks such as shape deformation,shape partial matching,skeleton extraction,shape correspondence,shape annotation and texture *** approaches have attempted to provide better segmentation solutions;however,the majority of the previous techniques used handcrafted features,which are usually focused on a particular attribute of 3Dobjects and so are difficult to *** this paper,we propose a three-stage approach for using Multi-view recurrent neural network to automatically segment a 3D shape into visually meaningful *** first stage involves normalizing and scaling a 3D model to fit within the unit sphere and rendering the object into different *** viewpoints,on the other hand,might not have been associated,and a 3D region could correlate into totally distinct outcomes depending on the *** address this,we ran each view through(shared weights)CNN and Bolster block in order to create a probability boundary *** Bolster block simulates the area relationships between different views,which helps to improve and refine the *** stage two,the feature maps generated in the previous step are correlated using a Recurrent Neural network to obtain compatible fine detail responses for each ***,a layer that is fully connected is used to return coherent edges,which are then back project to 3D objects to produce the final *** on the Princeton Segmentation Benchmark dataset show that our proposed method is effective for mesh segmentation tasks.

关键词： Deep learning mesh segmentation 3D shape shape features

来源：评论

学校读者我要写书评

暂无评论

A survey of UAV hardware selection

引用

Journal of engineering and Applied Science 2025年第1期72卷 1-21页

作者： Sdoukou, Evanthia Milidonis, Athanasios Efstathiou, Konstantinos Voyiatzis, Ioannis Department of Informatics and Computer Engineering University of West Attica Athens Greece

Applications concerning unmanned aerial vehicles (UAVs) have increased in recent years, mainly due to simple configuration, user-friendly navigation, and small weight and size. They are utilized and adopted within many contexts and services, such as mapping, construction inspection and surveillance, and search and rescue (SAR), changing the business perspective. Among different types of UAVs, multicopters or rotary-wing aircraft are the most popular, based on their vertical take-off and landing (VTOL) abilities, maneuverability, low cost, and easy deployment. Although there are many multicopter models available on the market, the prospect of building a custom one is becoming increasingly attractive since there is a variety of electronic parts to choose from. However, given that certain operations require special equipment for the effective operation of the UAV, the selection of the appropriate hardware parts is not an easy task. In this paper, a survey is presented concerning multicopter basic parts. For these parts, the most important methods are presented for calculating critical parameters in order for the total system to operate efficiently according to requirements. Finally, for critical application areas, a set of basic components is suggested for the multicopters’ efficient usage.

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：