检索结果-内蒙古大学图书馆

Cluster-based gradient method for stochastic optimal control problems with elliptic partial differential equation constraint

引用

NUMERICAL METHODS FOR PARTIAL DIFFERENTIAL EQUATIONS 2022年第6期38卷 1861-1879页

作者： Xiong, Meixin Chen, Liuhong Ming, Ju Hou, Lisheng Huazhong Univ Sci & Technol Sch Math & Stat Wuhan Peoples R China Iowa State Univ Dept Math Ames IA USA

In this article, we establish a cluster-based gradient method (CGM) by combining K-means clustering algorithm and stochastic gradient descent (SGD). By clustering the sampling solutions, we use the cluster centroids to represent sampling data and give an estimate to the full gradient. It is well known that the full gradient descent (FGD) can provide the steepest descent direction for finding a local minimum of the desired stochastic control problems. However, the huge computational requirements, which is proportional to the product of sample size and the numerical cost for each sample, often makes FGD cost prohibitive for large scale optimization problems. To reduce the formidable cost and the risks of getting stuck in a local minimum, SGD is proposed and can be regarded as a stochastic approximation of FGD. This, however, would result in a slow convergence due to the incorrect approximation of the iteratively update parameters. Our study shows that CGM could provide a good stochastic approximation to the full gradient with small sample size while has a more stable and faster convergence than SGD. To verify our algorithm, a stochastic elliptic control problem is selected and tested. The numerical results validate our method as a reliable gradient descent method with great potential applications in optimization problems.

关键词： clustering algorithm full gradient descent Monte Carlo method optimization under uncertainty PDE-constrained optimization stochastic gradient descent

来源：评论

学校读者我要写书评

暂无评论

New perspective of learning objects in e-learning system

引用

INTERNATIONAL JOURNAL OF INFORMATION AND LEARNING TECHNOLOGY 2023年第3期40卷 269-279页

作者： Amane, Meryem Aissaoui, Karima Berrada, Mohammed Sidi Mohamed Ben Abdellah Univ Fes Morocco Mohammed Premier Univ Oujda Morocco

PurposeTogether, learning objects (LOs) and e-pedagogical practices have the potential to improve the performance of e-learning systems in several ways. They can make e-learning more personalised and adaptable, providing students with a more engaging and effective learning ***/methodology/approachThe development of LOs and e-pedagogical practices have significantly influenced and changed the performance of e-learning systems. LOs are self-contained, reusable units of instructional content that create instructional materials, such as online courses, tutorials and assessments. They provide a flexible and modular approach to designing and delivering e-learning content, allowing educators to easily customise and adapt their materials to the needs of their students. e-pedagogical practices refer to the use of technology to enhance and support the teaching and learning process. They include strategies such as online collaboration, gamification and adaptive learning to improve student engagement, motivation and *** achieve this objective, this study consists of two main phases. First, the authors extract metadata from LOs using latent semantic analysis algorithms, which are considered a strong tool in web-mining exploration techniques. Second, they identify LOs according to a particular form of similarity using fuzzy c-means (FCM) algorithms. To improve classification accuracy, the FCM is used as a clustering ***/valueFinally, in order to assess the effectiveness of LOs with FCM, a series of experimental studies using a real-world dataset are conducted. The results of this study indicate that the proposed approach exceeds the traditional approach and produces good results.

关键词： Learning objects Web-based mining techniques Fuzzy c-means clustering algorithm Machine learning Latent semantic analysis (LSA)

来源：评论

学校读者我要写书评

暂无评论

High-density pedestrian detection algorithm based on deep information fusion

引用

APPLIED INTELLIGENCE 2022年第13期52卷 15483-15495页

作者： Zhang, Hexiang Yang, Xiaofang Hu, Ziyu Hao, Ruoxin Gao, Zehang Wang, Jianhao Yanshan Univ Qinhuangdao 066004 Hebei Peoples R China

In order to improve the accuracy of high-density population detection, a high density pedestrian detection algorithm (YOLOv4-HDPD) is proposed based on deep information fusion. By increasing the connection points of cross-layer fusion, high-level semantic information is further integrated with feature information. The improved Iterative Self-Organizing Data Analysis algorithm (ISODATA) makes the anchor value more suitable for the network model without increasing the number of parameters. Moreover, the network anti-interference ability is increased by replacing the CIOU algorithm target detection object. Compared with the original network, the YOLOv4-HDPD network has improved in mAP and avgIOU. Under the premise that the detection speed of the network is basically not affected, mAP is increased by 5.28% and avgIOU is increased by 5.73%. In terms of the current results, the network algorithm has been improved the detection effect of high-density pedestrians. At the same time, the network provides a new idea for solving the clustering and detection of dense targets in real scenes.

关键词： clustering algorithm Deep information fusion High density Pedestrian detection

来源：评论

学校读者我要写书评

暂无评论

Assessment of landslide susceptibility using DBSCAN-AHD and LD-EV methods

引用

Journal of Mountain Science 2022年第1期19卷 184-197页

作者： MAO Yi-min MWAKAPESA Deborah S LI Yi-can XU Kai-bin NANEHKARAN Yaser A ZHANG Mao-sheng School of Information Engineering Jiangxi University of Science and TechnologyGanzhou 341000China Key Laboratory for Geo-hazards in Loess Area Xi’an 710054China Northwest Nonferrous Geological Mining Group Limited Company Xi’an 710054China

Landslide susceptibility assessment plays a vital role in understanding landslide information in advance and taking preventive as well as control measures. The need to initially specify the number of clusters, difficulty in handling noise and quantifying rainfall data, limits the application of traditional clustering models in landslide susceptibility assessment which then lowers their performance accuracy. Thus, to overcome these limitations, this study proposed an improved clustering algorithm titled the DBSCAN-AHD algorithm, which combines the traditional DBSCAN(Density Based Spatial clustering of Application with Noise) algorithm and an Adaptive Hausdorff Distance(AHD) for landslide susceptibility modeling. Firstly, AHD was introduced to the traditional DBSCAN to quantify rainfall. Then,the DBSCAN-AHD grouped the mapping units with similar topology and geology characteristics into subclasses without specifying the number of clusters in advance for its ability to handle noise in the ***, the LD-EV(landslide density or eigenvalues) approach was introduced to obtain the susceptibility levels using K-means algorithm. Finally,to verify the model’s performance, statistical indices and the area under the curve(AUC) were applied and compared to the traditional DBSCAN, KPSO, Kmeans and Hierarchical clustering algorithms,whereby, the proposed model outperformed the others. Also, the obtained susceptibility map can provide references in taking relevant preventive and control measures.

关键词： Landslide susceptibility DBSCAN Machine learning algorithm Uncertain data clustering algorithm

来源：评论

学校读者我要写书评

暂无评论

Self-evaluation clustering scheme for modulation format identification in coherent optical communication

引用

OPTICAL FIBER TECHNOLOGY 2023年 75卷

作者： Zhang, Wenbo Ye, Jinmei Yue, Zixian Xu, Hengying Zhang, Xulun Sun, Peng Zhang, Xiaoguang Xi, Lixia Beijing Univ Posts & Telecommun Sch Sci Beijing 100876 Peoples R China Beijing Univ Posts & Telecommun Sch Elect Engn Beijing 100876 Peoples R China Beijing Univ Posts & Telecommun State Key Lab Informat Photon & Opt Commun Beijing 100876 Peoples R China Liaocheng Univ Sch Phys Sci & Informat Engn Liaocheng 252000 Peoples R China

A self-evaluation clustering scheme for modulation format identification in coherent optical communication is proposed. The performance of the scheme is tested by numerical simulations on a 28 Gbaud polarization division multiplexing (PDM) system. The results show that the proposed scheme achieves an identification accuracy of 100 % and has a significant tolerance to residual chromatic dispersion for PDM-4QAM, PDM-16QAM, PDM-32QAM, and PDM-64QAM signals when the values of optical signal-to-noise ratio are lower than the corre-sponding 7 % forward error correction thresholds. Moreover, the 20 Gbaud fiber transmission experiments of PDM-4QAM / 16QAM / 32QAM signals along the 2000 km, 1000 km, and 400 km standard single-mode fiber channel were also performed, respectively. The results show that the proposed scheme is robust towards long-haul fiber transmission.

关键词： Polarization division multiplexing system Modulation format identification clustering algorithm Residual chromatic dispersion

来源：评论

学校读者我要写书评

暂无评论

Data Stream clustering: An In-depth Empirical Study

引用

Proceedings of the ACM on Management of Data 2023年第2期1卷 1-26页

作者： Xin Wang Zhengru Wang Zhenyu Wu Shuhao Zhang Xuanhua Shi Li Lu Ohio State University Columbus OH USA Nvidia Shanghai China University of Manchester Manchester United Kingdom Singapore University of Technology and Design Singapore Singapore Huazhong University of Science and Technology Wuhan China Sichuan University Chengdu China

Data Stream clustering (DSC) plays an important role in mining continuous and unlabeled data streams in real-world applications. Over the last decades, numerous DSC algorithms have been proposed with promising clustering accuracy and efficiency. Despite the significant differences among existing DSC algorithms, they are commonly built around four key design aspects: summarizing data structure, window model, outlier detection mechanism, and offline refinement strategy. However, there is a lack of empirical studies on these key design aspects in the same codebase using real-world workloads with distinct characteristics. As a result, it is difficult for researchers to improve upon the state-of-the-art. In this paper, we conduct such a study of DSC on its four key design aspects. We implemented state-of-the-art variants of all of these design choices in an open-sourced platform from scratch and evaluated them using both real-world and synthetic workloads. Our analysis identifies the fundamental issues and trade-offs of each design choice in terms of both accuracy and efficiency. We even find that combining flexible design choices led to the development of a new algorithm called Benne, which can be tuned to achieve either better accuracy or better efficiency compared to the state-of-the-art.

关键词： clustering algorithm data stream empirical study

来源：评论

学校读者我要写书评

暂无评论

Improving Method of Anomaly Detection Performance for Industrial IoT Environment

引用

Computers, Materials & Continua 2022年第9期72卷 5377-5394页

作者： Junwon Kim Jiho Shin Ki-Woong Park Jung Taek Seo Department of Information Security Engineering Gachon UniversitySeongnam13120Korea Police Science Institute Korea National Police UniversityAsan31539Korea SysCore Lab. Sejong UniversitySeoul05006Korea Department of Computer Engineering Gachon UniversitySeongnam13120Korea

Industrial Control System(ICS),which is based on Industrial IoT(IIoT),has an intelligent mobile environment that supports various mobility,but there is a limit to relying only on the physical security of the ICS *** to various threat factors that can disrupt the workflow of the IIoT,machine learning-based anomaly detection technologies are being presented;it is also essential to study for increasing detection performance to minimize model errors for promoting stable ICS *** this paper,we established the requirements for improving the anomaly detection performance in the IIoT-based ICS environment by analyzing the related *** that,we presented an improving method of the performance of a machine learning model specialized for IIoT-based ICS,which increases the detection rate by applying correlation coefficients and clustering;it provides a mechanism to predict thresholds on a ***,we adopted the HAI dataset environment that actively reflected the characteristics of IIoT-based ICS and demonstrated that performance could be improved through comparative experiments with the traditional method and our proposed *** presented method can further improve the performance of commonly applied error-based detection techniques and includes a primary method that can be enhanced over existing detection techniques by analyzing correlation coefficients between features to consider feedback between ICS *** can contribute to improving the performance of several detection models applied in ICS and other areas.

关键词： Industrial IoT industrial control system anomaly detection clustering algorithm correlation coefficient

来源：评论

学校读者我要写书评

暂无评论

Major moisture source patterns for extreme precipitation events over the Chinese Loess Plateau

引用

INTERNATIONAL JOURNAL OF CLIMATOLOGY 2022年第15期42卷 7951-7967页

作者： Zhang, Mingxi Zhong, Deyu Huang, Wenyu Zhang, Yu Wang, Guangqian Li, Tiejian Tian, Yinglin Xie, Di Tsinghua Univ Dept Hydraul Engn State Key Lab Hydrosci & Engn Beijing Peoples R China Qinghai Univ Sch Water Resources & Elect Power Joint Sponsored State Key Lab Plateau Ecol & Agr Xining Peoples R China Tsinghua Univ Dept Earth Syst Sci Key Lab Earth Syst Modeling Minist Educ Beijing Peoples R China

Extreme precipitation (EP) has significant hydrological and environmental influences on the Chinese Loess Plateau (CLP) during the rainy season (June-September). In this study, we analysed the moisture sources for the EP events over the CLP during the rainy season. Three major moisture source patterns were identified through analysing the moisture source variability across the EP events with a newly developed method combining the Lagrangian moisture diagnostic and a clustering algorithm. South China and north China were determined as the most critical moisture source regions for the EP events over the CLP. The three moisture source patterns corresponded to three EP types, including the CLP-A type with a high contribution from north China (23.55%, the relative moisture contribution) and a relatively low contribution from south China (13.77%), and the CLP-B and CLP-C types, which had limited contributions (10.64 and 13.28%) from north China in common, and exhibited moderately high (19.30%) and extremely high (30.59%) contributions from south China, respectively. The moisture transport pathways for the three EP types were quantitatively analysed in detail. The favourable circulation patterns for the three EP types were further illustrated to reveal the atmospheric conditions for the spatial variations in moisture source patterns. In addition, this study provides a new strategy to analyse the moisture source variability, especially when it is difficult to find an easily manipulated criterion to classify the precipitation events.

关键词： Chinese Loess Plateau clustering algorithm extreme precipitation event Lagrangian moisture diagnostic moisture source

来源：评论

学校读者我要写书评

暂无评论

Research on the Machine Learning Platform Design Scheme Based on the Spark Platform 1st

Research on the Machine Learning Platform Design Scheme Base...

引用

World Conference on Intelligent and 3-D Technologies (WCI3DT)

作者： Zhong, Jianxu Yu, Shaofeng Xi, Lingzhi Ma, Yining CSG Power Generat Co Informat Commun Branch Guangzhou Guangdong Peoples R China

ISBN: (纸本)9789811971860;9789811971846;9789811971839

In the continuous innovation of computer technology, artificial intelligence technology has played a positive role in industrial production and daily life. Especially with the advent of the era of big data, cloud computing and Internet of Things technology has attracted people's attention. Using Spark to build a machine learning platform has become the main direction for guests and scholars to explore and analyze. In the study of practical construction, the machine learning platform with Spark as the core should be built with a clear grasp of distributed cluster technology and cloud computing, so as to flexibly use existing computing resources and improve the processing efficiency of practical data. At the same time, experimental comparative analysis should be carried out under the conditions of one-click environment and Hadoop, which proves that the hybrid architecture machine learning platform constructed in this paper has strong application performance.

关键词： Spark platform Machine learning Linear regression Platform measurement clustering algorithm

来源：评论

学校读者我要写书评

暂无评论

AI-enabled cluster head selection through modified density based clustering in Aeronautical Ad Hoc Networks✩

引用

AD HOC NETWORKS 2023年 148卷

作者： Shahbazi, Mohsen Simsek, Murat Kantarci, Burak Univ Ottawa Sch Elect Engn & Comp Sci 800 King Edward Ave Ottawa ON K1N 6N5 Canada

Offering in-flight connectivity has emerged as an essential demand for everyday flights. Such aircraft's high speed and dynamic characteristics make this task challenging, particularly in distant areas with no ground-to-air link. Aeronautical Ad Hoc Network (AANET), a network of commercial airplanes with air-to-air connectivity, is a viable solution to this problem. However, the instability of air-to-air links results in poor performance of such ad hoc networks. A cluster-based topology formation is an initial step to boost the performance and connectivity in such a network. Selecting a well-connected cluster head is the next stage in enhancing connection and stability. This study presents a new cluster head selection technique for AANETs that calculates the neighbor nodes within a given distance of each node and selects the node with the most connections as the new cluster head. A gateway node is chosen to facilitate connections with other clusters. According to simulations, the proposed method increases packet delivery ratio by 3%, end-to-end delay by 9%, and throughput by up to 10%. In addition, the proposed method reduces cluster head replacements by 17% and increases cluster head longevity by 8%.

关键词： In-flight connectivity AANET Ad hoc network AANET clustering clustering algorithm Cluster head selection Self organizing maps

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：