检索结果-南通市图书馆

Robust Recommendation Algorithm Based on Kernel Principal Component Analysis and Fuzzy C-means Clustering

在线全文

学校读者我要写书评

暂无评论

Wuhan University Journal of Natural Sciences 2018年第2期23卷 111-119页

作者： yi huawei NIU Zaiseng ZHANG Fuzhi LI Xiaohui WANG Yajun School of Electronics and Information Engineering Liaoning University of Technology Jinzhou 121001 Liaoning China School of Information Science and Engineering YanshanUniversity Qinhuangdao 066004 Hebei China

The existing recommendation algorithms have lower robustness in facing of shilling attacks. Considering this problem, we present a robust recommendation algorithm based on kernel principal component analysis and fuzzy c-means clustering. Firstly, we use kernel principal component analysis method to reduce the dimensionality of the original rating matrix, which can extract the effective features of users and items. Then, according to the dimension-reduced rating matrix and the high correlation characteristic between attack profiles, we use fuzzy c-means clustering method to cluster user profiles, which can realize the effective separation of genuine profiles and attack profiles. Finally, we construct an indicator function based on the attack detection results to decrease the influence of attack profiles on the recommendation, and incorporate it into the matrix factorization technology to design the corresponding robust recommendation algorithm. Experiment results indicate that the proposed algorithm is superior to the existing methods in both recommendation accuracy and robustness.

关键词： robust recommendation shilling attacks matrixfactorization kernel principal component analysis fuzzy c-meansclustering

Non-synchronous vibration of rotor blade in a six-stage transonic compressor

在线全文

学校读者我要写书评

暂无评论

Chinese Journal of Aeronautics 2024年第8期37卷 36-48页

作者： Ronghui CHENG Zhuo WANG huawei YU Lin DU yi ZHANG Xiaofeng SUN Aeroengine Research Institute Beihang UniversityBeijing 102206China School of Energy and Power Engineering Beihang UniversityBeijing 102206China

This paper presents an experimental study on the Non-Synchronous Vibration(NSV)in a six-stage transonic *** first part of the paper describes the NSV phenomenon of Rotor 1,which occurs when both Stator 1(S1)and Stator 2(S2)or S1 only are *** measurements and analysis are carried out for the former case through the unsteady wall pressure and the Blade Strain(BS).The spinning mode theory used in the rotor/stator interaction noise is employed to explain the relation between the circumferential wave number of the aerodynamic disturbance and the Nodal Diameter(ND)of the blade *** variations of the vibration amplitudes of different blades and the Inter-Blade Phase Angles(IBPAs)at different moments suggest that the evolution of NSV is a highly nonuniform phenomenon along the circumferential *** addition,the difference between the wall-pressure spectra generated by the NSV and the classic flutter has been *** the second part,the variations of aerodynamic loading due to the adjustment of the staggers of the Inlet Guide Vane(IGV),S1 and S2 have been *** is found that closing S1 only can result in a great fluctuation to the performance of the front stages,which might be detrimental to the flow organization and increase the risk of *** contrast,the effect of closing S2 only on the performance of the first two stages appears to be slighter relatively.

关键词： Transonic compressor Non-synchronous vibration Modal analysis Aerodynamic loading Nodal diameter

A Robust Collaborative Recommendation Algorithm Based on k-distance and Tukey M-estimator

维普期刊数据库评论

在线全文

维普期刊数据库

学校读者我要写书评

暂无评论

China Communications 2014年第9期11卷 112-123页

作者： yi huawei ZHANG Fuzhi LAN Jie School of Information Science and Engineering Yanshan University Qinhuangdao 066004 Hebei Province E R.China The Key Laboratory for Computer Virtual Technology and System Integration of Hebei Province Qinhuangdao 066004 P. R. China Liaoning University of Technology Jinzhou 121001 Liaoning Province P. R. China

The existing collaborative recommendation algorithms have lower robustness against shilling *** this problem in mind,in this paper we propose a robust collaborative recommendation algorithm based on k-distance and Tukey ***,we propose a k-distancebased method to compute user suspicion degree(USD).The reliable neighbor model can be constructed through incorporating the user suspicion degree into user neighbor *** influence of attack profiles on the recommendation results is reduced through adjusting similarities among ***,Tukey M-estimator is introduced to construct robust matrix factorization model,which can realize the robust estimation of user feature matrix and item feature matrix and reduce the influence of attack profiles on item feature ***,a robust collaborative recommendation algorithm is devised by combining the reliable neighbor model and robust matrix factorization *** results show that the proposed algorithm outperforms the existing methods in terms of both recommendation accuracy and robustness.

关键词： shilling attacks robust collaborative recommendation matrix factori-zation k-distance Tukey M-estimator

在线全文

学校读者我要写书评

暂无评论

基于改进蚁群算法的机械手三维操作路径规划

计算机应用与软件 2014年第4期31卷 302-304,307页

作者：伊华伟辽宁工业大学计算中心辽宁锦州121001

提出一种改进的蚁群算法,实现机械手在三维栅格模型中进行最优操作路径规划。利用该蚁群算法模拟蚂蚁的觅食过程,使蚂蚁在初始点和目标点之间采用"惯性原则"和最大信息素启发策略完成最优操作路径的搜索。仿真实验验证了该方法在机械手... 详细信息

提出一种改进的蚁群算法,实现机械手在三维栅格模型中进行最优操作路径规划。利用该蚁群算法模拟蚂蚁的觅食过程,使蚂蚁在初始点和目标点之间采用"惯性原则"和最大信息素启发策略完成最优操作路径的搜索。仿真实验验证了该方法在机械手操作路径规划应用中的有效性。即使在复杂的操作环境中,该方法也可以迅速地规划出最优操作路径。

关键词：三维操作空间三维栅格模型改进的蚁群算法惯性原则

Analysis of Clinical Characteristics and Risk Factors of Pseudomonas aeruginosa Bloodstream Infection in 55 Cases

在线全文

学校读者我要写书评

暂无评论

Yangtze Medicine 2023年第3期7卷 162-170页

作者： Pengpeng Tian Shanshan Su Tian Wang Lisa Zhu Wen Fan huawei yi Department of Laboratory Medicine of the First Affiliated Hospital of Yangtze University Jingzhou China

Objective: Pseudomonas aeruginosa bloodstream infection presents a severe challenge to hospitalized patients. To investigate the clinical characteristics, risk factors and drug resistance of Pseudomonas aeruginosa bloodstream infection. Methods: Clinical data and laboratory results of patients with Pseudomonas aeruginosa bloodstream infection in the First Affiliated Hospital of Yangtze University from January 2019 to December 2022 were retrospectively analyzed. The factors associated with infection and death were analyzed by univariate analysis. Results: A total of 55 patients were enrolled in this study, The 28-day mortality rate was 14.5%. Univariate analysis showed that high procalcitonin, low albumin, ICU admission, central venous catheterization, indwelling catheter, and mechanical ventilation were associated with death. Multivariate Logistic regression analysis showed that hypoproteinemia and central venous catheters were independent risk factors for death in patients with Pseudomonas aeruginosa bloodstream infection. Conclusions: The drug resistance of P. aeruginosa bloodstream infection is not high, but the fatality rate is high. The combination of hypoalbuminemia after the onset of the disease and the use of central vein catheters can lead to increased mortality, suggesting that clinical identification of high-risk patients as early as possible, reducing the use of catheters, preventing the occurrence of P. aeruginosa bloodstream infection and improving the prognosis.

关键词： Pseudomonas aeruginosa Bloodstream Infections Resistance Risk Factors

Layer-by-layer epitaxy growth of thickness-controllable two-dimensional tungsten disulfide

维普期刊数据库评论

在线全文

维普期刊数据库

学校读者我要写书评

暂无评论

Science China(Information Sciences) 2024年第5期67卷 302-310页

作者： Jieyuan LIANG Zixing ZOU Junwu LIANG Di WANG Biao WANG Anshi CHU Jiali yi Cheng ZHANG Lizhen FANG Tian ZHANG huawei LIU Xiaoli ZHU Dong LI Anlian PAN Key Laboratory for Micro-Nano Physics and Technology of Hunan Province State Key Laboratory of Chemo/Biosensing and Chemometrics Hunan Institute of Optoelectronic IntegrationCollege of Materials Science and Engineering School of Physics and ElectronicsHunan University School of Electronic Information and Electrical Engineering Changsha University School of Physics and Telecommunication Engineering Yulin Normal University Units Semiconductor Technology Co. Ltd. School of Physics and Electronics Hunan Normal University

Bilayer transition metal dichalcogenides（TMDs） balance the high mobility of single layers with the high state density of multilayers and therefore have promising application prospects in high-performance electronics. However, the layer-controlled growth of 2D materials is still confronted with challenges such as poor repeatability between different labs and a limited understanding of the growth mechanism at the atomic scale. Herein, we report a new carbon-assisted chemical vapor deposition process that can realize the growth of WS2sheets with high yield, precise thickness controllability, and repeatability. We show that carbon can act as a reducing agent and catalyst that preferentially reacts with the WO3precursor to form intermediate WO3-xproducts with low-valence state W. The resulting oxycarbide gas has a low surface adsorption energy when deposited on the surface of as-grown WS2, which provides nucleation sites for the subsequent layer of WS2growth and leads to the vertical growth of WS2sheets. The growth mechanism is thoroughly investigated. Electrical transport measurements show that the produced bilayer WS2possesses a high carrier mobility(up to 58 cm2·V-1·s-1) and small subthreshold swing（estimated to be 148 mV/decade）,which are among the best reported results for TMDs produced using CVD.

关键词： WS$_2$ thickness-controlled growth chemical vapor deposition additive carbon

基于模糊核聚类和支持向量机的鲁棒协同推荐算法

在线全文

同方期刊数据库

学校读者我要写书评

暂无评论

电子与信息学报 2017年第8期39卷 1942-1949页

作者：伊华伟张付志巢进波燕山大学信息科学与工程学院秦皇岛066004 河北省计算机虚拟技术与系统集成重点实验室(燕山大学) 秦皇岛066004 辽宁工业大学电子与信息工程学院锦州121001

该文针对现有推荐算法在面对托攻击时鲁棒性不高的问题,提出一种基于模糊核聚类和支持向量机的鲁棒推荐算法。首先,根据攻击概貌间高度相关的特性,利用模糊核聚类方法在高维特征空间对用户概貌进行聚类,实现攻击概貌的第1阶段检测。然后... 详细信息

该文针对现有推荐算法在面对托攻击时鲁棒性不高的问题,提出一种基于模糊核聚类和支持向量机的鲁棒推荐算法。首先,根据攻击概貌间高度相关的特性,利用模糊核聚类方法在高维特征空间对用户概貌进行聚类,实现攻击概貌的第1阶段检测。然后,利用支持向量机分类器对含有攻击概貌的聚类进行分类,实现攻击概貌的第2阶段检测。最后,基于攻击概貌检测结果,通过构造指示函数排除攻击概貌在推荐过程中产生的影响,并引入矩阵分解技术设计相应的鲁棒协同推荐算法。实验结果表明,与现有的基于矩阵分解模型的推荐算法相比,所提算法不但具有很好的鲁棒性,而且准确性也有提高。

关键词：鲁棒推荐算法托攻击矩阵分解模糊核聚类支持向量机

在线全文

学校读者我要写书评

暂无评论

通过包络面重构的大规模粒子并行绘制算法

国防科技大学学报 2024年第5期46卷 219-227页

作者：王华维艾志玮曹轶北京应用物理与计算数学研究所计算物理重点实验室北京100088 中物院高性能数值模拟软件中心北京100088

针对大规模粒子高表现可视化需求,提出基于包络面重构的大规模粒子并行绘制算法。该算法以连续曲面的形式表示,绘制大规模粒子的团簇表面及其物理量分布。对算法进行了分布式并行化,从而可以通过大规模并行来处理亿以上规模的粒子数据... 详细信息

针对大规模粒子高表现可视化需求,提出基于包络面重构的大规模粒子并行绘制算法。该算法以连续曲面的形式表示,绘制大规模粒子的团簇表面及其物理量分布。对算法进行了分布式并行化,从而可以通过大规模并行来处理亿以上规模的粒子数据。在算法实现上,还解决了并行计算时的块间裂缝问题,并提出了快速查找邻域粒子的方法,同时,基于可见性对粒子数据进行剔除,提高了绘制效率。由此,可以通过带光照效果的光滑曲面来高表现展示大规模粒子数据中的团簇结构及其物理量分布。实验结果表明,该算法在512核上可在5 s内完成上亿粒子的绘制,并行效率可达60%。该算法已成功应用到大规模并行非平衡分子动力学模拟等实际模拟应用中。

关键词：粒子可视化包络面距离场分布式并行化可见性剔除

在线全文

学校读者我要写书评

暂无评论

推荐系统的隐私保护研究综述

计算机科学与探索 2023年第8期17卷 1814-1832页

作者：冯晗伊华伟李晓会李锐辽宁工业大学电子与信息工程学院辽宁锦州121001

推荐系统需要大规模地提取相关用户的历史数据信息作为预测模型的训练集,用户提供的数据量越大、越具体,个人信息越容易被推断出来,致使个人隐私的泄露,从而使用户对服务提供商的信任程度降低,不再向系统提供相关数据,导致系统推荐精度... 详细信息

推荐系统需要大规模地提取相关用户的历史数据信息作为预测模型的训练集,用户提供的数据量越大、越具体,个人信息越容易被推断出来,致使个人隐私的泄露,从而使用户对服务提供商的信任程度降低,不再向系统提供相关数据,导致系统推荐精度降低甚至较难完成推荐。因此,如何能在对用户隐私进行保护的前提下,获取用户信息进行准确性高的有效推荐已成为当前的一个研究热点。首先,对隐私保护技术进行概述,主要包括差分隐私技术、同态加密技术、联邦学习技术和安全多方计算技术,并对这几种常用的隐私保护技术进行了比较;然后,从平衡隐私注入和推荐精度之间的关系角度出发介绍客户端、服务器端和客户-服务器端所采用的隐私保护技术,系统阐述了国内外推荐系统隐私保护的研究成果,并在此基础上进行总结、对比和分析;接下来,对比基于差分隐私技术的推荐算法的实验效果,并分析相应技术存在的不足;最后,对基于隐私保护的推荐系统的未来发展方向提出相应的建议和展望。

关键词：推荐系统隐私保护推荐方式隐私技术