检索结果-内蒙古大学图书馆

3rd International Conference on Machine Learning, Cloud computing and intelligent Mining, MLCCIM 2024

作者： Chen, Chuanxu Li, Ruiting Guan, Quansheng Shi, Zhan Zeng, Ying Zheng, Jianping School of Electronic and Information Engineering South China University of Technology Guangzhou China Electric Power Dispatch and Control Center Guangdong Power Grid Co. Ltd. Guangzhou China

ISBN: (纸本)9789819616978

The Lyapunov drift-based scheduling algorithms control the queue length to stabilize the system and maximize the throughput for multi-resource multi-queue cloud/edge systems. However, these algorithms often ignore the heterogeneity in job length, thus leading to a long delay for job processing. To this end, this paper takes the job completion time (JCT) as a penalty in the drift-plus-penalty expression to minimize the average job competition time, and proposes a Lyapunov drift-plus-time (LDPT) based scheduling algorithm for multi-resource queuing computing systems. The optimality of LDPT with respect to average JCT and queue length is analyzed. By using the Amazon EC2 to simulate, the results show that LDPT can improve average JCT and average queuing time fairness. © The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd. 2025.

关键词： Scheduling algorithms

来源：评论

学校读者我要写书评

暂无评论

A Lightweight Knowledge Distillation and Feature Compression Model for User Click-Through Rates Prediction in Edge computing Scenarios

引用

IEEE INTERNET OF THINGS JOURNAL 2025年第3期12卷 2295-2308页

作者： Yang, Bin Zhou, Jiawei Zhang, Shihao Xing, Ying Jiang, Weiwei Xu, Lexi China Unicom Res Inst Graph Neural Network Beijing 100048 Peoples R China China Unicom Res Inst Artificial Intelligence Team Beijing 100048 Peoples R China China Unicom Res Inst Network Intelligent Operat Res Ctr Beijing Peoples R China Beijing Univ Posts & Telecommun Sch Artificial Intelligence Beijing 100876 Peoples R China Beijing Univ Posts & Telecommun Sch Informat & Commun Engn Key Lab Universal Wireless Commun Minist Educ Beijing 100876 Peoples R China

Along with the development of Internet of Things systems, numerous edge intelligent devices can obtain a large amount of user data, and the analysis of this user data can be applied to business scenarios, such as user click-through rate (CTR) prediction. In the recommendation, advertising and other scenarios, the users at the edge have high response requirements for CTR prediction model training and inference. In the current edge scenario of CTR prediction, there are problems of overly complex model structure and highly sparse original features, which makes it difficult to deploy CTR prediction models at the edge. Therefore, we propose KD-based graph attention FI model (KD-GAFIM), a lightweight recommendation algorithm that combines graph neural networks (GNNs) with knowledge distillation (KD). The approach uses graph attention networks to flexibly capture feature dependencies in a way that maintains a small model size while augmenting the feature vector with feature dependencies. And by sharing the embedding layer of the teacher model, KD-GAFIM improves the efficiency of user CTR prediction. On top of that, we also propose a feature compression strategy guided by model interpretability, which identifies high-contributing features for inference and model refinement based on their performance in model interpretability. This strategy improves efficiency, making KD-GAFIM suitable for training and inference on edge devices. We conducted extensive experiments on multiple datasets. The experimental results show that KD-GAFIM outperforms various state-of-the-art CTR prediction models, demonstrating that GNN-based KD models can improve model performance while reducing model size and feature dimensionality, and have significant potential for application at the edge.

关键词： Predictive models Computational modeling Data models Vectors Analytical models Accuracy Performance evaluation Training Recommender systems Frequency modulation Click-through rate (CTR) prediction edge learning feature compression Internet of Things (IoT) systems knowledge distillation (KD)

来源：评论

学校读者我要写书评

暂无评论

A novel fractional order grey Euler model and its application in clean energy prediction

引用

ENERGY 2025年 322卷

作者： Yang, Zhongsen Wang, Yong Fan, Neng Wen, Shixiong Kuang, Wenyu Yang, Mou Sapnken, Flavian Emmanuel Narayanan, Govindasami Li, Hong-Li Southwest Petr Univ Sch Sci Chengdu 610500 Sichuan Peoples R China Sichuan Univ State Key Lab Intelligent Construction & Hlth Oper Chengdu 610065 Sichuan Peoples R China Wuyi Univ Key Lab Cognit Comp & Intelligent Informat Proc Fu Jiangmen 529020 Guangdong Peoples R China Southwest Petr Univ Natl Key Lab Oil & Gas Reservoir Geol & Exploitat Chengdu 610500 Sichuan Peoples R China Univ Douala Univ Inst Technol Lab Technol & Appl Sci POB 8698 Douala Cameroon Kunsan Natl Univ Sch IT Informat & Control Engn 588 Daehak Ro Gunsan Si 54150 Jeonbuk South Korea Xinjiang Univ Coll Math & Syst Sci Urumqi 830017 Xinjiang Peoples R China

The global imperative for sustainable energy solutions has intensified due to escalating energy demands and heightened environmental regulations, making clean energy adoption critical for strategic policymaking. This study introduces an advanced fractional grey Euler model (FEGEM(p,1)) incorporating a novel fractional-order accumulated generating operation with exponential kernel functionality (r-EAGO) to forecast clean energy production trends in hydroelectricity, nuclear power, and natural gas. The proposed r-EAGO mechanism demonstrates superior capability in extracting implicit patterns from historical data through its sophisticated accumulation process. The fractional-order modelling framework offers enhanced flexibility through its adjustable differential order, particularly effective in addressing the inherent non-linearity and complexity of clean energy systems. To ensure optimal performance, the model hyperparameters are optimised using a differential evolution algorithm. The results of Monte-Carlo simulation and probability density analysis reveal good robustness of the model. Comprehensive validation of the seven forecasting models revealed that the FEGEM(p,1) model is extremely accurate, achieving a mean absolute percentage error (MAPE) of 2.508 %, 3.489 %, and 2.343 % for the three energy cases, respectively. Projections of the next five years indicate distinct growth trajectories: Nuclear power generation and natural gas production are forecasted to maintain compound annual growth rates of 4.65 % and 5.46 % respectively through 2027, while hydropower exhibits cyclical growth patterns influenced by hydrological variations, ultimately reaching 1,429 billion kWh by the terminal forecast year. Finally, these findings yield strategic recommendations for optimising energy development framework.

关键词： Grey model Differential evolution algorithm Monte-Carlo simulation Clean energy production

来源：评论

学校读者我要写书评

暂无评论

Research on dynamic network security detection technology based on SDN 4

Research on dynamic network security detection technology ba...

引用

4th International Conference on Network Communication and Information Security, ICNCIS 2024

作者： Zhao, Kai Beijing Polytechnic Beijing100176 China

ISBN: (数字)9781510688261

ISBN: (纸本)9781510688254

Traditional network architecture can no longer meet the development needs of technologies such as cloud computing and big data. SDN network architecture has the characteristics of high openness and programmability, which can quickly respond to changes in business requirements. SDN decouples the control management function of the network from the data forwarding function, simplifies network configuration work based on logic centralized controllers and open programming interfaces, thereby achieving a flat management mode and flexible data forwarding function. The centralized control and programmability of SDN networks also pose significant security threats to SDN. Based on the CSE-CIC-IDS2018 dataset, this article tests the performance of various intelligent learning algorithms, analyzes data flows, distinguishes normal data from abnormal data, improves detection accuracy, and has practical significance for the security defense of SDN networks. © 2025 SPIE.

关键词： Network security Computer security Network architectures control systems Machine learning Data transmission Defense and security Detection and tracking algorithms Information security Security technologies

来源：评论

学校读者我要写书评

暂无评论

A dual vision-guided mobile robot control approach for multi-target path planning and intelligent pickup

引用

ROBOTICS AND AUTONOMOUS systems 2025年 190卷

作者： Zhang, Wenyong Xia, Dawen Chang, Guoyan Hu, Yang Huo, Yujia Wang, Ziqiang Feng, Fujian Zheng, Zhiquan Li, Yantao Li, Huaqing Guizhou Minzu Univ Coll Data Sci & Informat Engn Guiyang 550025 Peoples R China South China Univ Technol Sch Comp Sci & Engn Guangzhou 510006 Peoples R China Kaili Univ Engn Res Ctr Micronano & Intelligent Mfg Minist Educ Kaili 556011 Peoples R China Guizhou Med Univ Affiliated Hosp Guiyang 550001 Peoples R China Guizhou Univ State Key Lab Publ Big Data Guiyang 550025 Peoples R China Chongqing Univ Coll Comp Sci Chongqing 400044 Peoples R China Southwest Univ Coll Elect & Informat Engn Chongqing 400715 Peoples R China

With the rapid development of the Internet of Robotic Things (IoRT), the autonomous sensing and decision-making capabilities of IoRT in multi-target pickup tasks have become a research hotspot. However, existing methods still face significant challenges in optimizing multi-target path planning and intelligent pickup operations. To this end, we propose a dual vision-guided mobile robot control approach based on the IoRT that combines multi-target path planning and intelligent pickup, called MTPP, in which robots employ dual cameras as sensors to pick up multiple targets. Specifically, the MTPP approach consists of an ant colony optimization algorithm which considers the rotation angle and the target distance (AD-ACO), and the dual-camera intelligent pickup method (DIP). In the AD-ACO algorithm, we present a new method to calculate the actual distance between targets based on the pixel coordinates of two points (PDC). Then, we convert the robot's rotating angle when pickup each target into the corresponding arc length of rotation. Next, the actual distance and the arc length of rotation are summed to obtain the effective distance. The effective distance is employed as a parameter of the ant colony optimization algorithm for path planning to generate the optimal path. Moreover, the DIP method is composed of a pickup robot, a robot positioning method, and a dual-camera navigation method, where the robot is utilized for collecting targets, the positioning method is used for real-time robot localization, and the dual-camera navigation method is adopted to guide the robot to the targets. Finally, our MTPP approach is applied to a smart tennis collection system enabling efficient and accurate operations.

关键词： Internet of Robotic Things (IoRT) Mobile robot positioning Mobile robot navigation Multi-target path planning intelligent pickup

来源：评论

学校读者我要写书评

暂无评论

QSM: A Quick Shuffling Method for Defending Against Low-Rate Denial of Service Attacks in Microservice systems

QSM: A Quick Shuffling Method for Defending Against Low-Rate...

引用

2024 International Conference on Mechatronics and intelligent control, ICMIC 2024

作者： Zhao, Zheng Fan, Qiqi Zhao, Chunchang Zhao, Qi Xie, Xin Dalian Maritime University Dalian China Northeastern University Shenyang China Hunan University of Information Technology Changsha China Hunan Province Higher Education Key Laboratory of Intelligent Sensing and Computing Changsha China

ISBN: (数字)9781510686847

ISBN: (纸本)9781510686830

Microservice architecture, renowned for its flexible scalability and loose coupling, has seen widespread adoption. However, this architecture has also become a prime target for network attackers, with low-rate denial of service (LDoS) attacks posing significant threats to microservice systems. Due to their covert nature, traditional passive defense methods often fall short in countering LDoS attacks. To address this challenge, this paper introduces a microservice-oriented quick shuffling method (QSM), based on the moving target defense (MTD) principle. The QSM framework leverages the dynamic and flexible properties of microservices to implement a rapid shuffling mechanism, ensuring the protection of innocent clients. Experimental results demonstrate the effectiveness of QSM in robustly defending microservice systems against LDoS attacks. © 2025 SPIE.

关键词： Denial-of-service attack

来源：评论

学校读者我要写书评

暂无评论

UDS for Data Communication in Temperature Monitoring 3rd

UDS for Data Communication in Temperature Monitoring

引用

3rd International Conference on intelligent computing Techniques for Smart Energy systems, ICTSES 2023

作者： Mallikarjun, B.C. Prathima, H.S. Department of ETE Siddaganga Institute of Technology Karnataka Tumkur India

ISBN: (纸本)9789819784288

Electronic control units (ECUs) are computer systems which are capable to process this data and send signals to the various systems in the vehicle to modify their behavior in order to optimize performance and ensure safety. Maintaining appropriate operating temperatures within ECUs is essential for preventing overheating, which can lead to system malfunction or failure. Overheating can cause the components of an ECU to degrade over time, leading to decreased performance, increased energy consumption, and potential system failure. This proposed work provides a temperature monitoring solution, i.e., Unified Diagnostic Service (UDS), that enhances the overall functionality of ECUs and contributes to vehicle safety and performance. The temperature monitoring solution would enable the ECUs to measure and monitor the temperature of the vehicles components and systems, which would help to reduce the risk of malfunctions and overheating, which can lead to catastrophic failures and safety risks. © The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd. 2025.

关键词： Catastrophic Electronic control units Temperature monitoring

来源：评论

学校读者我要写书评

暂无评论

Research on New Railway Master Data Management Technology Based on Artificial Intelligence and Blockchain

Research on New Railway Master Data Management Technology Ba...

引用

International Conference on Artificial Intelligence and Autonomous Transportation, AIAT 2024

作者： Sun, Siqi Zhang, Kai Zou, Dan Ma, Xiaoning Wang, Peiran Wu, Jiang Institute of Computing Technologies Chinese Academy of Railway Sciences Haidian District Beijing100081 China

ISBN: (纸本)9789819639687

Railway master data is crucial for ensuring consistency and accuracy across information systems. As railway informatization progresses, the types and volumes of data have rapidly increased, making traditional master data management methods increasingly inadequate. On the other hand, the railway industry has high demands for data processing speed and security. Therefore, effectively managing high-value data that is reused across different business domains is key to improving the efficiency and safety of railway systems. This study proposes a new master data management technology by combining advanced artificial intelligence (AI) and blockchain technology to address the pain points in railway master data management. Experiments show that applying this framework can intelligently identify and classify complex business data, effectively supporting dynamic updates and intelligent governance of master data. © Beijing Paike Culture Commu. Co., Ltd. 2025.

关键词： Data consistency

来源：评论

学校读者我要写书评

暂无评论

An Overview of the Use of Artificial Intelligence Techniques to Improve Power Quality 3rd

An Overview of the Use of Artificial Intelligence Techniques...

引用

3rd International Conference on intelligent systems and Sustainable computing, ICISSC 2023

作者： Baggam, Swathi Rani, R. Sheba School of Engineering Malla Reddy University Hyderabad India

ISBN: (纸本)9789819783540

An extensive analysis of the use of artificial intelligence (AI) approaches to enhance power quality in electrical grids is presented in this paper. In maintaining the power quality there might be many problems like frequency variations harmonic distortions and voltage instability. Conventional approaches to these problems are unable to handle the dynamic and intricate structure of contemporary electrical grids. By facilitating predictive modelling, real-time monitoring, and adaptive control, artificial intelligence (AI) approaches including machine learning, neural networks, and optimization algorithms present intriguing answers. In addition to classifying AI-based methods for improving power quality, this assessment summarises the body of research in the area and highlights problems, trends, and potential future possibility enhancement. © The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd. 2025.

关键词： Power grids

来源：评论

学校读者我要写书评

暂无评论

Integration of Perovskite/Low-Dimensional Material Heterostructures for Optoelectronics and Artificial Visual systems

引用

ADVANCED FUNCTIONAL MATERIALS 2025年

作者： Du, Yu-Jin Yang, Junjie Lv, Ziyu Zhai, Yongbiao Yi, Zezhuang Xie, Yuhui Zheng, Ming-Lin Ma, Xiangyu Gong, Guodong Wang, Yan Zhou, Ye Han, Su-Ting Shenzhen Univ Coll Elect & Informat Engn Shenzhen 518060 Peoples R China Shenzhen Univ Inst Microscale Optoelect Shenzhen 518060 Peoples R China Hefei Univ Technol Sch Microelect Hefei 230009 Peoples R China Shenzhen Univ Inst Adv Study Shenzhen 518060 Peoples R China Hong Kong Polytech Univ Dept Appl Biol Kowloon Hong Kong 999077 Peoples R China Hong Kong Polytech Univ Chem Technol & Res Inst Smart Energ Kowloon Hong Kong 999077 Peoples R China

Heterojunctions combining halide perovskites with low-dimensional materials are revolutionizing optoelectronic device design by leveraging complementary properties. Halide perovskites, known for their tunable bandgaps, excellent light-harvesting, and efficient charge carrier mobility, provide a robust foundation for photodetectors (PDs) and imaging sensors. Low-dimensional materials contribute ultrafast carrier mobility, enhanced light-matter interactions, and mechanical flexibility. When integrated into heterostructures, these materials enable precise control over charge dynamics, leading to significant improvements in device efficiency, stability, and response speed. This synergy addresses critical challenges in optoelectronics, advancing flexible electronics, wearable sensors, and high-sensitivity imaging systems. Ongoing advancements in interface engineering and material synthesis are continually enhancing the reliability and operational efficacy of these devices across various environmental conditions. Additionally, these heterostructures show substantial promise in neuromorphic computing, where their optoelectronic properties support energy-efficient, event-driven data processing. By mimicking the adaptive and hierarchical nature of biological visual systems, they offer new possibilities for real-time image analysis and intelligent decision-making. This review highlights the latest developments in halide perovskite-based heterojunctions with low-dimensional materials and their transformative role in bridging the gap between artificial and biological vision, driving advancements in technologies such as adaptive robotics and bio-inspired visual systems.

关键词： halide perovskites heterojunctions low-dimensional materials memory neuromorphic computing

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：