检索结果-内蒙古大学图书馆

Winne-Leader-Follower a Novel Hit Allocation Algorithm for Pixel Detectors

IEEE TRANSACTIONS ON NUCLEAR SCIENCE 2024年第5期71卷 1250-1256页

作者： Jirsa, J. Marcisovsky, M. Janoska, Z. Jakovenko, J. Czech Tech Univ Fac Elect Engn Dept Microelect Prague 16627 Czech Republic Czech Tech Univ Fac Nucl Sci & Phys Engn Dept Phys Prague 11519 Czech Republic

The integration of front-end electronics into pixel electronics enables hybrid pixel detectors to capture a high-resolution X-ray image. The decreasing size of the technology node allows pixel electronics to integrate in tens of micrometers of pixel pitch. On the other hand, the decreasing pixel pitch size leads to new problems, such as the charge-sharing effect and fluorescent photons in high-Z materials. These effects decrease the spatial and spectroscopic resolution of the detectors. This article presents a novel asynchronous hit allocation algorithm, Winner-Leader-Follower (WLF). The proposed algorithm operates in the digital domain and performs clustering and hit allocation within $5 \times 5$ pixels. The algorithm's concept and implementation are presented, followed by measurements obtained with a test chip designed in a 65-nm complementary metal-oxide semiconductor process.

关键词： clustering algorithms Photonics Resource management Detectors Fluorescence X-ray imaging System-on-chip Digital integrated circuits position-sensitive particle detectors radiation detector circuits semiconductor radiation detectors X-ray detectors

来源：评论

学校读者我要写书评

暂无评论

Anomaly Detection of Big Data Based on Improved Fast Density Peak clustering Algorithm 5th

Anomaly Detection of Big Data Based on Improved Fast Density...

引用

5th EAI International Conference on Multimedia Technology and Enhanced Learning, ICMTEL 2023

作者： Zhong, Fulong Lin, Tongxi Guangzhou Huashang Vocational College Guangzhou511300 China

ISBN: (纸本)9783031505768

Aiming at the problems of traditional clustering algorithms in the process of large amount of big data anomaly monitoring under the background of big data, a big data anomaly detection method based on improved fast density peak clustering algorithm is proposed. Set up a big data anomaly clustering framework, automatically select parameters and clustering centers, evaluate big data outliers through standardized local density and distance, and can obtain outliers, extract and calculate thresholds according to features, complete anomaly scheduling, and achieve anomaly detection. The clustering algorithm designed in this paper is used for example analysis, which shows that the algorithm designed in this paper can meet the needs of actual users and improve the detection accuracy of power big data outliers. © ICST Institute for Computer Sciences, Social Informatics and Telecommunications Engineering 2024.

关键词： clustering algorithms

来源：评论

学校读者我要写书评

暂无评论

DBKNN and Radial-ANFIS Model for Energy Efficient Wireless Sensor Network

引用

IEEE ACCESS 2024年 12卷 15917-15929页

作者： Rhesa, M. J. Revathi, S. Vellore Inst Technol Sch Elect Engn Vellore 632014 India

In a Wireless Sensor Network (WSN), packet transmission and sensing functions are the most energy consumption factors. When it comes to wireless communication, unnecessary sensing could increase data communication which in turn needs the biggest amount of energy for communication. Thus, to minimize Energy Consumption (EC), various research has been conducted. Still, they don't provide efficient EC. To solve this issue, a novel Davies-Bouldin K Nearest Neighbor (DBKNN) and Radial-Adaptive Neuro Fuzzy Interference System (Radial-ANFIS) model is proposed for energy-efficient WSNs. Initially, partitioning is performed on deployed nodes. Then, node features are extracted. By employing Laplacian Cubic Cheetah Optimizer (LCCO), CHs are selected using features;then, nodes are clustered using DBKNN. Next, weights are assigned to each clustered node. By employing Adaptive Intensed Golden Tortoise Beetle Optimizer (AI-GTBO), relay nodes are selected using weight value. Paths are created betwixt the node and Base Station (BS) utilizing relay nodes;then, the shortest path is computed. Nodes sense the data once the network is designed, and the correlation between each data is calculated. For redundant node detection, the analytic output is given to Radial-ANFIS. Lastly, redundant nodes are kept to sleep, and the remaining node data are utilized for transmission. When data is transmitted to BS, its first dimension is reduced using Newton Raphson Iterative Principal Component Analysis (NRI-PCA). According to outcomes, the proposed model reduces EC and maximizes the overall throughput and network lifetime compared to other methods. The EC rate is also reduced to 1364 J using the proposed method.

关键词： Wireless sensor networks Energy efficiency Feature extraction Routing Relays clustering algorithms Protocols Artificial intelligence Nearest neighbor methods AI-GTBO Davies-Bouldin K Nearest Neighbor Laplacian Cubic Cheetah Optimizer NRI-PCA Radial-ANFIS

来源：评论

学校读者我要写书评

暂无评论

Optimum Partition of Power Networks Using Singular Value Decomposition and Affinity Propagation

引用

IEEE TRANSACTIONS ON POWER SYSTEMS 2024年第5期39卷 6359-6371页

作者： Eddin, Maymouna Ez Massaoudi, Mohamed Abu-Rub, Haitham Shadmand, Mohammad Abdallah, Mohamed Texas A&M Univ Dept Elect Engn College Stn TX 77840 USA Texas A&M Univ Dept Elect Engn Doha 23874 Qatar Univ Illinois Dept Elect & Comp Engn Chicago IL 60607 USA Hamad Bin Khalifa Univ Coll Sci & Engn Div Informat & Comp Technol Doha 34110 Qatar

Due to coupling and correlation between nodes and buses in the power system, Power Grid Partitioning (PGP) is a promising approach to analyze large power systems and provide timely actions during disturbances. From this perspective, this paper proposes an efficient framework for fast and optimal PGP, based on singular value decomposition analysis of the graph's Laplacian. An Affinity Propagation clustering algorithm-based PGP is tailored for automatically forming highly interconnected clusters based on pairwise similarities without requiring a predefined number of partitions. The core objective is to quantify the clustering performance based on internal clustering validity indices, such as the Silhouette Index, Calinski-Harabasz Index, and Davies-Bouldin Index. The adopted methodology aims to enhance partitioning efficiency substantially while preserving a high level of partitioning quality. The proposed framework is verified on IEEE 14, 39, 118, and 2000-bus systems and compared to nine other well-known and widely used clustering techniques, including K-Means and Gaussian Mixture models. The simulation results demonstrate the scalability of the proposed approach and its high-quality partitioning output with a Silhouette index of 0.6162, 0.6597, 0.6664, and 0.6555 for the IEEE 14, 39, 118, and 2000-bus systems, respectively.

关键词： Indexes clustering algorithms Partitioning algorithms Power systems Power grids Scalability Laplace equations complex networks grid partitioning machine learning power system analysis

来源：评论

学校读者我要写书评

暂无评论

Automated Correction of GIS Data for Loads and Distributed Energy Resources in Secondary Distribution Networks

引用

IEEE TRANSACTIONS ON POWER SYSTEMS 2024年第1期39卷 1628-1636页

作者： Montano-Martinez, Karen Ma, Shanshan Vittal, Vijay Rojas, Cynthia Arizona State Univ Sch Elect Comp & Energy Engn Tempe AZ 85287 USA Arizona Publ Serv Phoenix AZ 85072 USA

Currently, utilities are making significant efforts to correct their secondary models in geographic information systems (GISs) to capture the distribution system performance due to the increased penetration of distributed energy resources (DERs) such as renewable energy resources, distributed energy storage, and electric vehicles (EVs). However, their databases are often not reliable in terms of GIS coordinates of elements such as loads and DERs resulting in an approximate representation of the secondary distribution network. This paper presents an automated tool for accurate secondary network topology GIS coordinates correction. The proposed automated tool provides a more accurate feeder topology for utilities to estimate and operate distribution systems by assigning the load and DER nodes to their corresponding customer location. To simplify the complexity of the proposed new tool, only two commonly available inputs are being used as input data: municipal parcel GIS delimitation data, and utility secondary feeder topology database. A three-stage framework is proposed to develop the automated tool: the first stage reads and processes the raw input data;the second stage works automatically with no human intervention to assign the load and DER nodes to their associated location;the third stage provides the load and DER coordinates and physical address. The proposed procedure is implemented on an actual feeder from an electrical utility with high penetration of *** results show that the proposed method provides a more accurate estimation of the topology of the feeder in terms of coordinates.

关键词： Databases Geographic information systems Topology clustering algorithms Network topology Load modeling Data processing Distribution system distribution system modeling geographic information system (GIS) secondary system modeling smart grids

来源：评论

学校读者我要写书评

暂无评论

FIG: Feature-Weighted Information Granules With High Consistency Rate

引用

IEEE TRANSACTIONS ON BIG DATA 2024年第4期10卷 400-414页

作者： Cai, Jianghe Deng, Yuhui Zhou, Yi Huang, Jiande Min, Geyong Jinan Univ Dept Comp Sci Guangzhou 510632 Peoples R China Columbus State Univ TSYS Sch Comp Sci Columbus GA 31907 USA Univ Exeter Coll Engn Math & Phys Sci Dept Comp Sci Exeter EX44QF England

Information granules are effective in revealing the structure of data. Therefore, it is a common practice in data mining to use information granules for classifying datasets. In the existing granular classifiers, the information granules are often classified according to the standard membership function only without considering the influence of different feature weights on the quality of granules and label classification results. In this article, we utilize the feature weighting of data to produce the information granules with high consistency rate called FIG. First, we use consistency rate and contribution scores to generate information granules. Then, we propose a granular two-stage classifier GTC based on FIG. GTC divides the data into fuzzy and fixed points and then calculates the interval matching degree to assign data points to the most suitable cluster in the second step. Finally, we compare FIG with two state-of-the-art granular models (T-GrM and FGC-rule), and classification accuracy is also compared with other classification algorithms. The extensive experiments on synthetic datasets and public datasets from UCI show that FIG has sufficient performance to describe the data structure and excellent capability under the constructed granular classifier GTC. Compared with T-GrM and FGC-rule, the time overhead required for FIG to obtain information granules is reduced by an average of 51.07%, the per unit quality of the granules is also increased by more than 14.74%. Compared with other classification algorithms, an average of 5.04% improves GTC accuracy.

关键词： clustering algorithms Classification algorithms Granular computing Data models Data mining Computational modeling Standards Data modeling information granule feature weighting consistency classifier

来源：评论

学校读者我要写书评

暂无评论

SLO-Aware Function Placement for Serverless Workflows With Layer-Wise Memory Sharing

引用

IEEE TRANSACTIONS ON PARALLEL AND DISTRIBUTED SYSTEMS 2024年第6期35卷 919-936页

作者： Cheng, Dazhao Yan, Kai Cai, Xinquan Gong, Yili Hu, Chuang Wuhan Univ Sch Comp Sci Wuhan 430072 Hubei Peoples R China

Function-as-a-Service (FaaS) is a promising cloud computing model known for its scalability and elasticity. In various application domains, FaaS workflows have been widely adopted to manage user requests and complete computational tasks efficiently. Motivated by the fact that function containers collaboratively use the image layer's memory, co-placing functions would leverage memory sharing to reduce cluster memory footprint, this article studies layer-wise memory sharing for serverless functions. We find that overwhelming memory sharing by placing containers in the same cluster machine may lead to performance deterioration and Service Level Objective (SLO) violations due to the increased CPU pressure. We investigate how to maximally reduce cluster memory footprint via layer-wise memory sharing for serverless workflows while guaranteeing their SLO. First, we study the container memory sharing problem under serverless workflows with a static Directed Acyclic Graph (DAG) structure. We prove it is NP-Hard and propose a 2-approximation algorithm, namely MDP. Then we consider workflows with dynamic DAG structure scenarios, where the memory sharing problem is also NP-Hard. We design a Greedy-based algorithm called GSP to address this issue. We implement a carefully designed prototype on the OpenWhisk platform, and our evaluation results demonstrate that both MDP and GSP achieve a balanced and satisfying state, effectively reducing up to 63$\%$% of cache memory usage while guaranteeing serverless workflow SLO.

关键词： Containers Memory management Heuristic algorithms clustering algorithms Cache memory Scalability Directed acyclic graph Serverless computing serverless cluster container placement scheduling

来源：评论

学校读者我要写书评

暂无评论

Block-Level Knowledge Transfer for Evolutionary Multitask Optimization

引用

IEEE TRANSACTIONS ON CYBERNETICS 2024年第1期54卷 558-571页

作者： Jiang, Yi Zhan, Zhi-Hui Tan, Kay Chen Zhang, Jun South China Univ Technol Sch Comp Sci & Engn Guangzhou 510006 Peoples R China Hong Kong Polytech Univ Dept Comp Hong Kong Peoples R China Zhejiang Normal Univ Key Lab Intelligent Educ Technol & Applicat Zhejia Jinhua 321017 Peoples R China Hanyang Univ ERICA Ansan 15588 South Korea

Evolutionary multitask optimization is an emerging research topic that aims to solve multiple tasks simultaneously. A general challenge in solving multitask optimization problems (MTOPs) is how to effectively transfer common knowledge between/among tasks. However, knowledge transfer in existing algorithms generally has two limitations. First, knowledge is only transferred between the aligned dimensions of different tasks rather than between similar or related dimensions. Second, the knowledge transfer among the related dimensions belonging to the same task is ignored. To overcome these two limitations, this article proposes an interesting and efficient idea that divides individuals into multiple blocks and transfers knowledge at the block-level, called the block-level knowledge transfer (BLKT) framework. BLKT divides the individuals of all the tasks into multiple blocks to obtain a block-based population, where each block corresponds to several consecutive dimensions. Similar blocks coming from either the same task or different tasks are grouped into the same cluster to evolve. In this way, BLKT enables the transfer of knowledge between similar dimensions that are originally either aligned or unaligned or belong to either the same task or different tasks, which is more rational. Extensive experiments conducted on CEC17 and CEC22 MTOP benchmarks, a new and more challenging compositive MTOP test suite, and real-world MTOPs all show that the performance of BLKT-based differential evolution (BLKT-DE) is superior to the compared state-of-the-art algorithms. In addition, another interesting finding is that the BLKT-DE is also promising in solving single-task global optimization problems, achieving competitive performance with some state-of-the-art algorithms.

关键词： Task analysis Knowledge transfer Optimization Statistics Sociology clustering algorithms Benchmark testing Block-level knowledge transfer (BLKT) differential evolution (DE) evolutionary computation (EC) evolutionary multitask optimization (EMTO)

来源：评论

学校读者我要写书评

暂无评论

Adversarial Robustness on Image Classification With k-Means

引用

IEEE ACCESS 2024年 12卷 28853-28859页

作者： Omari, Rollin Kim, Junae Montague, Paul Australian Dept Def Def Sci & Technol Grp Edinburgh SA 5111 Australia

Attacks and defences in adversarial machine learning literature have primarily focused on supervised learning. However, it remains an open question whether existing methods and strategies can be adapted to unsupervised learning approaches. In this paper we explore the challenges and strategies in attacking a k -means clustering algorithm and in enhancing its robustness against adversarial manipulations. We evaluate the vulnerability of clustering algorithms to adversarial attacks on two datasets (MNIST and Fashion-MNIST), emphasising the associated security risks. Our study investigates the impact of incremental attack strength on training, introduces the concept of transferability between supervised and unsupervised models, and highlights the sensitivity of unsupervised models to sample distributions. We additionally introduce and evaluate an adversarial training method that improves testing performance in adversarial scenarios, and we highlight the importance of various parameters in the proposed training method, such as continuous learning, centroid initialisation, and adversarial step-count. Overall, our study emphasises the vulnerability of unsupervised learning and clustering algorithms to adversarial attacks and provides insights into potential defence mechanisms.

关键词： Training clustering algorithms Robustness Testing Mathematical models Unsupervised learning Sensitivity Adversarial machine learning Machine learning Adversarial examples adversarial machine learning adversarial robustness adversarial training k-means clustering unsupervised learning

来源：评论

学校读者我要写书评

暂无评论

Big Data Analytics on Lung Cancer Diagnosis Framework With Deep Learning

引用

IEEE-ACM TRANSACTIONS ON COMPUTATIONAL BIOLOGY AND BIOINFORMATICS 2024年第4期21卷 757-768页

作者： Guan, Peiyuan Yu, Keping Wei, Wei Tan, Yanlin Wu, Jia Jiangxi Univ Chinese Med Sch Comp Sci & Technol Nanchang 330004 Jiangxi Peoples R China Waseda Univ Global Informat & Telecommun Inst Tokyo 1698555 Japan Xian Univ Technol Sch Comp Sci & Engn Xian 710048 Peoples R China Cent South Univ Xiangya Hosp 2 PET CT Ctr Changsha 410083 Peoples R China Monash Univ Res Ctr Artificial Intelligence Clayton Vic 3800 Australia

As the segment of diseased tissue in PET images is time-consuming, laborious and low accuracy, this work proposes an automated framework for PET image screening, denoising and diseased tissue segmentation. First, taking into account the characteristics of PET images, the framework uses a differential activation filter to select whole-body images containing lesion tissue. Second, a new neural network containing residual connections which has powerful generalization performance compared with normal FCN network is proposed for PET image reconstruction and denoising. Finally, in the segmentation of lesion tissues, a custom clustering algorithm based on the density is used to distinguishe the lesion tissue part from the normal tissue. Tests on real medical PET images show that the whole automated framework has good performance and time cost in PET lesion image screening, image denoising and lesion tissue segmentation compared with other algorithms. The framework shows promising scientific study and application prospects.

关键词： Image segmentation Lesions Positron emission tomography Noise reduction Neural networks clustering algorithms Gaussian noise Deep learning image classification image noise reduction image segmentation cluster segmentation

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：