检索结果-内蒙古大学图书馆

STR: Spatio-temporal trajectory representation learning with dual-focus encoder for whole trajectory similarity computation

引用

Information Fusion 2025年 123卷

作者： Li, Mengqiu Niu, Xinzheng Zhu, Jiahui Fournier-Viger, Philippe Wu, Youxi School of Computer Science and Engineering University of Electronic Science and Technology of China Chengdu China school of Information and Software Engineering University of Electronic Science and Technology of China Chengdu China College of Computer Science and Software Engineering Shenzhen University Shenzhen China School of Artificial Intelligence Hebei University of Technology Tianjin China

In the field of trajectory data mining, trajectory similarity computation is a key issue associated with trajectory representation learning. Current methods of trajectory similarity computation often represent trajectory data of varying lengths as vectors with uniform dimension. However, existing trajectory representations fall short of comprehensively calculating trajectory similarity: first, they focus solely on spatial information, while overlooking the temporal information of the trajectory, and second, they represent the temporal and spatial information separately. To address these limitations, we propose a novel Spatio-temporal Trajectory Representation method, named STR, aimed at capturing both the features of individual trajectories and the relationships between them in a three-dimensional spatio-temporal region from the perspectives of both trajectory points and whole trajectories, and improving the accuracy of trajectory similarity metrics. STR consists of two main modules: (i) a trajectory feature tokenization module, which considers the uneven density distribution of trajectory points in the three-dimensional space to build a hierarchical structure that considers space and time as a whole, and extracts spatio-temporal features to construct a token sequence;(ii) a dual-focus trajectory encoder learns the features of a given trajectory and those of similar trajectories, thereby enabling the consideration of relationships between whole trajectories. Moreover, a density-based whole trajectory similarity metric is designed to find similar trajectories. Experiments were conducted on three public datasets, and the accuracy of the trajectory representations obtained with STR is significantly greater than for existing state-of-the-art baselines on a variety of distance metrics. © 2025 Elsevier B.V.

关键词： Dual-focus trajectory encoder Representation learning Spatio-temporal trajectory Trajectory similarity

来源：评论

学校读者我要写书评

暂无评论

In-Range Farthest Point Queries and Related Problem in High Dimensions 49

In-Range Farthest Point Queries and Related Problem in High ...

引用

49th EATCS International Conference on Automata, Languages, and Programming, ICALP 2022

作者： Huang, Ziyun Xu, Jinhui Department of Computer Science and Software Engineering Penn State Erie The Behrend College United States Department of Computer Science and Engineering State University of New York BuffaloNY United States

ISBN: (纸本)9783959772358

Range-aggregate query is an important type of queries with numerous applications. It aims to obtain some structural information (defined by an aggregate function F(·)) of the points (from a point set P) inside a given query range B. In this paper, we study the range-aggregate query problem in high dimensional space for two aggregate functions: (1) F(P ∩ B) is the farthest point in P ∩ B to a query point q in d and (2) F(P ∩ B) is the minimum enclosing ball (MEB) of P ∩ B. For problem (1), called In-Range Farthest Point (IFP) Query, we develop a bi-criteria approximation scheme: For any ϵ > 0 that specifies the approximation ratio of the farthest distance and any γ > 0 that measures the "fuzziness" of the query range, we show that it is possible to pre-process P into a data structure of size Õϵ,γ(dn1+ρ) in Õϵ,γ(dn1+ρ) time such that given any Rd query ball B and query point q, it outputs in Õϵ,γ(dnρ) time a point p that is a (1 − ϵ)-approximation of the farthest point to q among all points lying in a (1 + γ)-expansion B(1 + γ) of B, where 0 © Ziyun Huang and Jinhui Xu;licensed under Creative Commons License CC-BY 4.0

关键词： Aggregates

来源：评论

学校读者我要写书评

暂无评论

Smart Classroom: A Deep Learning Approach towards Attention Assessment through Class Behavior Detection

Smart Classroom: A Deep Learning Approach towards Attention ...

引用

2022 Advances in science and engineering Technology International Conferences, ASET 2022

作者： Parambil, Medha Mohan Ambali Ali, Luqman Alnajjar, Fady Gochoo, Munkhjargal United Arab Emirates University College of IT Department of Computer Science and Software Engineering Al Ain United Arab Emirates

ISBN: (纸本)9781665418010

The role of education in shaping future generations is crucial. The rapid advancement in Artificial Intelligence (AI) has prompted a considerable turn of events in the educational field. Advanced AI technologies have the potential to replace traditional classroom practices, giving more importance to students' preferences. Assessing the attention in students is the building block to improve both instructing and learning. Most of the time, it becomes puzzling for an instructor to track every student and know their interest in learning, especially in a vast scenario. In such cases, instructors are often unaware of the most coherent approach while teaching students with different learning capabilities. Therefore, this paper proposed a system that incorporates AI to develop a real-time vision-based smart classroom that autonomously monitors students' attention, as well as, emotions and gives live graphical feedback to the instructor. Our system was able to monitor the presence of a student in the classroom and analyze the students' attention and emotion. The proposed approach was able to assign numerical scores to the student for the concentration level in the class. The stated system can help the instructors evaluate the student's attention instantaneously and more accurately, thus supporting them in conducting the session suitably that may lead to better academic performance. © 2022 IEEE.

关键词： Students

来源：评论

学校读者我要写书评

暂无评论

Proactive Hybrid Autoscaling for Container-Based Edge Applications in Kubernetes 19th

Proactive Hybrid Autoscaling for Container-Based Edge Appli...

引用

19th EAI International Conference on Heterogeneous Networking for Quality, Reliability, Security and Robustness, QShine 2023

作者： Zhu, Kaile Shen, Shihao Lan, Shizhan Wang, Xiaofei Zhang, Cheng Qiu, Chao Leung, Victor College of Intelligence and Computing Tianjin University Tianjin China China Mobile Guangxi Institute Co. Ltd. Shanghai China Institute of Technology Tianjin University of Finance and Economics Tianjin China College of Computer Science and Software Engineering Shenzhen University Shenzhen China Department of Electrical and Computer Engineering The University of British Columbia Vancouver Canada

ISBN: (纸本)9783031651229

As the rising of the Internet of Things (IoT), edge computing is widely adopted in numerous applications. However, current autoscaling tools are not designed for edge applications and can not utilize the heterogeneous resources of edge nodes efficiently. In this paper, we propose a proactive hybrid autoscaler specifically optimized for edge computing scenario. With the Bidirectional Long Short Term Memory (Bi-LSTM) based load prediction model, the proposed autoscaler is able to predict the future workload and perform scaling operation before it arrives. In addition, a overload compensation algorithm is implemented to mitigate the Quality of Service (QoS) decreasing due to under-prediction. Then, a hybrid scaling method is applied to simultaneously modify the number of pods and their resource quotas without restarting. Experimental results with a real-world workload dataset shows the proposed load prediction model has better accuracy compared with the Long Short Term Memory model and the state-of-the-art statistical analysis model, Autoregressive Integrated Moving Average (ARIMA), which is also more than 350 times slower than our model in prediction speed. Finally, evaluation in a real Kubernetes cluster shows that the proposed proactive hybrid autoscaler outperforms the default Horizontal Pod Autoscaler (HPA) of Kubernetes in terms of both QoS and resource utilization efficiency. © ICST Institute for computer sciences, Social Informatics and Telecommunications engineering 2024.

关键词： Quality of service

来源：评论

学校读者我要写书评

暂无评论

Review of Power Grid Optimization Scheduling Methods Based On Deep Reinforcement Learning

Review of Power Grid Optimization Scheduling Methods Based O...

引用

China International Electrical and Energy Conference (CIEEC)

作者： Yang Zherui Zhang Zhen Li Fangzhou Dai Nan Yan Yiying Wang Ke Dayu College Hohai University College of Energy and Electrical Engineering Hohai University College of Computer Science and Software Engineering Hohai University

ISBN: (数字)9798350359558

ISBN: (纸本)9798350359565

In recent years, with the introduction of energy conservation and emission reduction targets, the introduction of renewable energy has become an important initiative. However, due to the many factors that affect the new energy units, the traditional scheduling method becomes inappropriate. If a more intelligent scheduling method is not developed, the lack of flexible energy scheduling capacity will further lead to problems such as insufficient wind power consumption level, inefficient power grid decision-making, high operating costs, and unstable energy supply. In order to solve the above problems and improve the environmental protection, efficiency, economy and safety of the power grid, this paper summarizes the application of artificial intelligence in the field of power grid scheduling, analyzes SAC, A3C and TD3 algorithms in detail, analyzes their mechanism through the algorithm structure, and summarizes the advantages and disadvantages of each algorithm and its application scenarios. In addition, this paper also analyzes the application of artificial intelligence in unit combination, economic scheduling optimization and optimal power flow optimization. Different from other review papers, this paper deeply analyzes several problems that may arise in the application of the algorithm, and specifically analyzes the future optimization direction and corresponding optimization methods, so as to improve the level of experience combination, fault handling, comprehensive consideration, interpretability and other aspects of the future agent.

关键词： Training Renewable energy sources Reviews Wind power generation Deep reinforcement learning Power grids Scheduling

来源：评论

学校读者我要写书评

暂无评论

F-IKOS: An Abstract Interpretation-based Static Analyzer for Fortran Programs 12

F-IKOS: An Abstract Interpretation-based Static Analyzer for...

引用

12th International Workshop on Quantitative Approaches to software Quality, QuASoQ 2024

作者： Zou, Sheng Chen, Liqian Fan, Guangsheng Huang, Renjie Yin, Banghu College of Computer Science and Technology National University of Defense Technology Changsha410073 China State Key Laboratory of Complex & Critical Software Environment Changsha410073 China College of Systems Engineering National University of Defense Technology Changsha410073 China

The Fortran programming language is widely utilized in numerical computation and scientific computing. Fortran programs are prone to potential runtime errors related to numerical properties due to the large number of numerical operations. In this paper, we present F-IKOS, an abstract interpretation-based static analyzer for Fortran programs on top of IKOS, which soundly handles floating-point types in Fortran programs. Firstly, we translate Fortran programs to LLVM IR using compiler front-end Flang. After that, we extend IKOS to support sound floating-point analysis and then employ it to analyze the translated LLVM IR. Particularly, when analyzing floating-point types in programs, we first abstract floating-point expressions into real-number expressions with interval coefficients, and then linearize these expressions into real-number expressions with scalar coefficients. These linear expressions are subsequently handled by abstract domains originally designed for real-number types to produce sound analysis results. We have conducted experiments on representative Fortran programs to show the efficiency and effectiveness of F-IKOS. The experimental results are encouraging: F-IKOS soundly analyzes runtime errors in complex programs, outperforming other analyzers. © 2024 Copyright for this paper by its authors. Use permitted under Creative Commons License Attribution 4.0 International (CC BY 4.0).

关键词： FORTRAN (programming language)

来源：评论

学校读者我要写书评

暂无评论

Stratifying large software files to improve prediction performance in software defect prediction 9

Stratifying large software files to improve prediction perfo...

引用

9th ACIS International Virtual Conference on Applied Computing and Information Technology, ACIT 2022

作者： Alshehri, Yasser Ali Alnazzawi, Noha Hijazi, Haneen Alharbi, Rawan Computer Science and Engineering Dept Yanbu Industrial College Saudi Arabia Dept of Software Engrg Hashemite University Jordan

ISBN: (纸本)9781450397605

Size is one of the significant factors associated with bugs, and it has been used to predict software faults. We believe that stratifying software files based on size can play an essential role in improving prediction performance. This study explored the effect of size by stratifying our sample based on each unit's size and distributing software units in multiple stratified groups based on an equal distribution approach. We stratified the Eclipse Europa project files, and we reported the performance of each stratified group and compared them. We used two popular classifiers, decision tree J48, and random forest, to implement this experiment. These classifiers presented similar results on the same group of files. The results indicated that predicting faults with large files is better than predicting those in small files. In addition, the results showed higher median values of all performance measures and less variation in each measure. © 2022 ACM.

关键词： Decision trees

来源：评论

学校读者我要写书评

暂无评论

An Intelligent Defect Detection Algorithm for PCB based on Deep Learning 6

An Intelligent Defect Detection Algorithm for PCB based on D...

引用

6th International Conference on Advances in Image Processing, ICAIP 2022

作者： Zhu, Xiangyuan Xiao, Xiuchun Lan, Zhiming Hong, Qiuhong Hou, Miao School of Computer Science and Software Zhaoqing University China College of Electronics and Information Engineering Guangdong Ocean University China

ISBN: (纸本)9781450397155

As an essential component of modern machines, printed circuit board (PCB) is widely used in various electronic products. Its quality significantly affects the quality of products. However, the production process of PCB is often accompanied with defects. In this paper, a defect detection algorithm is proposed. Data augmentation such as flipping, shifting, brightness adjustment, rotation, and Guass noise are carried out to diversify the dataset. You only look once (YOLO) v5s is then introduced to detect the PCB defects. Through parameter tuning and optimization, a trained detection model is achieved. F1-score and mean average precision (mAP) are used to assess the performance of the model. The experiment results show that the mAP and F1-score are 99.3% and 99.0%, respectively. The model developed based on YOLO-v5s algorithm can achieve superior performance, which is competent to detect the defects of PCBs. © 2022 ACM.

关键词： Printed circuit boards

来源：评论

学校读者我要写书评

暂无评论

Hybrid Machine Learning Model for Face Recognition Using SVM

引用

computers, Materials & Continua 2022年第8期72卷 2697-2712页

作者： Anil Kumar Yadav R.K.Pateriya Nirmal Kumar Gupta Punit Gupta Dinesh Kumar Saini Mohammad Alahmadi IES College of Technology Bhopal462044India Maulana Azad National Institute of Technology Bhopal462003India Department of Information Technology Manipal University JaipurJaipur303007India Department of Computer and Communication Engineering Manipal University JaipurJaipur303007India Department of Software Engineering College of Computer Science and EngineeringUniversity of JeddahJeddah23218Saudi Arabia

Face recognition systems have enhanced human-computer interactions in the last ten ***,the literature reveals that current techniques used for identifying or verifying faces are not immune to *** Component Analysis-Support Vector Machine(PCA-SVM)and Principal Component Analysis-Artificial Neural Network(PCA-ANN)are among the relatively recent and powerful face analysis *** to PCA-ANN,PCA-SVM has demonstrated generalization capabilities in many tasks,including the ability to recognize objects with small or large data *** from requiring a minimal number of parameters in face detection,PCA-SVM minimizes generalization errors and avoids overfitting problems better than ***-SVM,however,is ineffective and inefficient in detecting human faces in cases in which there is poor lighting,long hair,or items covering the subject’s *** study proposes a novel PCASVM-based model to overcome the recognition problem of PCA-ANN and enhance face *** experimental results indicate that the proposed model provides a better face recognition outcome than PCA-SVM.

关键词： Face recognition system(FRS) face identification SVM discrete cosine transform(DCT) artificial neural network(ANN) machine learning

来源：评论

学校读者我要写书评

暂无评论

Multimodal News Recommendation Based on Deep Reinforcement Learning 7

Multimodal News Recommendation Based on Deep Reinforcement L...

引用

7th International Conference on Intelligent Computing and Signal Processing, ICSP 2022

作者： Guo, Nan Fu, Zhangpeng Zhao, Qihui Northeastern University School of Computer Science and Engineering Shenyang China Northeastern University School of Software College Shenyang China

ISBN: (纸本)9781665478571

Multimodal news recommendation is a challenging problem due to the rapid expansion of Internet information, bringing different levels of knowledge expression such as text, images, audio, and video, etc. In this paper, we propose a multimodal news recommendation method based on a deep reinforcement learning framework to represent user interests as multimodal information. The proposed method feeds the multimodal fusion feature into the rainbow agent of deep reinforcement learning to learn news representation. The experiment on the MIND and IM-MIND datasets give the result of AUC, MRR, nDCG@5, and nDCG@10 scores, which outperform LSTUR, FIM, and DKN, NRMS, NPA, and DeepFM models. It shows that reinforcement learning is an excellent choice to fulfill recommendation tasks, and the multimodal fusion feature is effective for learning accurate news representations. © 2022 IEEE.

关键词： Deep 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：