检索结果-内蒙古大学图书馆

A resistance outlier sampling algorithm for imbalanced data prediction

INTELLIGENT DATA ANALYSIS 2022年第3期26卷 583-598页

作者： Pan, Xiaoying Jia, Rong Huang, Jiahao Wang, Hao Xian Univ Posts & Telecommun Sch Comp Sci & Technol Xian Shaanxi Peoples R China Key Lab Network Data Anal & Intelligent Proc Xian Shaanxi Peoples R China Northwestern Polytech Univ Sch Software Xian Shaanxi Peoples R China

Classification of imbalanced data is an important challenge in current research. sampling is an important way to solve the problem of imbalanced data classification, but some traditional sampling algorithms are susceptible to outliers. Therefore, an iF-ADASYN sampling algorithm is proposed in this paper. First, based on the ADASYN algorithm, we introduce the isolation Forest algorithm to overcome its vulnerability to outliers. Then, a calculation method of anomaly index which can delete outliers accurately of minority data is presented. The experimental results of four UCI public imbalanced datasets show that the algorithm can effectively improve the accuracy of the minority class, and increase the stability. In the real thrombus dataset, the AUC value of the iF-ADASYN algorithm is more significant than that of SMOTE and ADASYN algorithms, and the recognition rate of patients with thrombosis increased by 20%. The iF-ADASYN algorithm obtains better resistance to outliers than the original ADASYN algorithm. Meanwhile, it improves the accuracy of minority class decision boundary region division.

关键词： Imbalanced data sampling algorithm outlier thrombosis risk prediction

来源：评论

学校读者我要写书评

暂无评论

Novel sampling algorithm for DSP controlled 2 kW PFC converter

引用

IEEE TRANSACTIONS ON POWER ELECTRONICS 2001年第2期16卷 217-222页

作者： Zhou, JH Lu, ZY Lin, ZY Ren, YC Qian, ZM Wang, YS Zhejiang Univ Elect Engn Coll Hangzhou 310027 Peoples R China

This paper proposes a novel sampling algorithm for digital signal processing (DSP) controlled 2 kW power factor correction (PFCs) converters, which can improve switching noise immunity greatly in average-current-control power supplies. Based on the newly developed DSP chip TMS320F240, a 2 kW PFC stage is implemented. The novel sampling algorithm shows great advantages when the converter operates at a frequency above 30 kHz.

关键词： digital signal processing power factor correction sampling algorithm

来源：评论

学校读者我要写书评

暂无评论

A Cautionary Note on the Hanurav-Vijayan sampling algorithm

引用

JOURNAL OF SURVEY STATISTICS AND METHODOLOGY 2022年第5期10卷 1276-1291页

作者： Chauvet, Guillaume Univ Rennes CNRS ENSAI IRMAR UMR 6625 F-35000 Rennes France

We consider the Hanurav-Vijayan (HV) sampling design, which is the default method programmed in the SURVEYSELECT procedure of the SAS software. We prove that it is equivalent to the Sunter procedure but is capable of handling any set of inclusion probabilities. We prove that the Horvitz-Thompson (HT) estimator is not generally consistent under this sampling design. We propose a conditional HT estimator and prove its consistency under a nonstandard assumption on the first-order inclusion probabilities. Since this assumption seems difficult to control in practice, we recommend not to use the HV sampling design.

关键词： Consistency sampling algorithm Sunter algorithm Unequal probability sampling

来源：评论

学校读者我要写书评

暂无评论

PVSA : A general and elegant sampling algorithm for Voxel-based 3D object detection 10

PVSA : A general and elegant sampling algorithm for Voxel-ba...

引用

10th International Conference on Control, Automation and Robotic (ICCAR)

作者： Gong, Diancheng Li, Junru Wang, Chunchun Wang, Zhiling Univ Sci & Technol China Chinese Acad Sci Heifei Inst Phys Sci Inst Intelligent Machines Hefei Peoples R China Anhui Univ Sci & Technol Chinese Acad Sci Heifei Inst Phys Sci Inst Intelligent Machines Hefei Peoples R China

ISBN: (纸本)9798350373172;9798350373189

Perceiving the environment is vital for autonomous vehicles as it serves as the foundation for decision making and path planning. LiDAR is a widely employed sensor, which produces a voluminous and sparsely populated point cloud. For voxel-based 3D object detection methods, the initial step involves the division of the raw point cloud into voxels, the process known as voxelization. Nevertheless, once the number of point clouds contained within a voxel reaches the certain threshold, the allocation of additional point clouds to that voxel ceases. This leads to a greater degree of information loss. Scholars primarily focus on the subsequent stages following voxelization, such as feature extraction and utilization. We first focus on the sampling issue during the voxelization. In the paper, we propose a general and elegant Points in Voxel sampling algorithm module named PVSA. During the voxelization, the assignment of all points into their respective voxels continues even after the maximum number of points in a voxel has been reached. For voxels in which the number of internal point clouds exceeds the certain threshold, the farthest distance sampling method is utilized as it ensures a genuine and uniform distribution of the point cloud within the voxel. We conducted an evaluation of the proposed module using the Kitti dataset. Experimental findings suggest that the incorporation of the PVSA module enhances the object detection capabilities of the voxel-based model, particularly in the identification of samll targets like pedestrians. The incorporation of PVSA modules significantly enhances Pillarnet's capacity to recognize pedestrians, resulting in a 46.2% pt improvement in performance at a distance of 20 meters. On average, there is an enhancement of 1.43% pt.

关键词： sampling algorithm Voxelization Small Object Detection Autonomous Vehicle

来源：评论

学校读者我要写书评

暂无评论

A New sampling algorithm used in MultiLevel UV Factorization method for Scattering from Random Rough Surface 1

A New Sampling Algorithm used in MultiLevel UV Factorization...

引用

IEEE International Conference on Computational Electromagnetics (ICCEM)

作者： Yu, Zherui Wang, Haogang Shi, Jiancheng He, Lingfang Zhejiang Univ Dept Informat Sci & Elect Rd 38th Yuquan Campus Hangzhou Zhejiang Peoples R China

ISBN: (纸本)9781479962815

In this paper, we propose a new sampling algorithm combined with multilevel UV(MLUV) factorization method to calculate the scattering from Gaussian random rough surface with exponential correlation function. The new sampling algorithm is based on the steepness of patch pairs which support the basis functions. The numerical analyses in this paper show that the proposed algorithm significantly improves the accuracy of the UV approximation of matrix Z(K).

关键词： MLUV rough surface scattering sampling algorithm

来源：评论

学校读者我要写书评

暂无评论

Frequency sampling algorithm applied in microwave measurements based on step-size control method 8

Frequency sampling algorithm applied in microwave measuremen...

引用

8th International Conference on Electronics, Computers and Artificial Intelligence (ECAI)

作者： Rosca, Cosmina Mihaela Paraschiv, Nicolae Petr Gas Univ Control Engn Comp & Elect Dept Ploiesti Romania

ISBN: (纸本)9781509020478

Extending the microwave frequency measurement from very low to very high frequency requires improvements for the acquisition time. In this paper, a new sampling algorithm is presented with the main purpose to reduce the acquisition time by using a limited number of samples. The adaptive algorithm proposed in this paper computes only a limited number of samples and then reconstructs the entire circuit response using the interpolation model. This method uses an adaptive step-size control and has the initial step and the error predefined. The algorithm evaluates the difference between two consecutive S parameters. The adaptive step-size algorithm assumes that, when the distance between the current S parameter value and the previous one decreases (below a threshold epsilon), the exploration step-size may be increased up to a limit in order to keep the step-size to a moderate value. Otherwise, it might overlook major S parameter variations. The biggest challenge here is represented by the correlation between S parameter domain, frequency domain and scaling values. This algorithm automatically finds the number of points needed to accurately evaluate high S parameters variations and it has a high speed computing.

关键词： microwave measurements vector network analyzer S parameters adaptive step - size sampling algorithm reduced number of samples

来源：评论

学校读者我要写书评

暂无评论

Frequency sampling algorithm applied in microwave measurements based on extreme points 10

Frequency sampling algorithm applied in microwave measuremen...

引用

10th International Conference on Electronics, Computers and Artificial Intelligence (ECAI)

作者： Rosca, Cosmina-Mihaela Paraschiv, Nicolae Petr Gas Univ Control Engn Comp & Elect Dept Ploiesti Romania

ISBN: (纸本)9781538649015

This paper presents a new frequency sampling algorithm with the main purpose to reduce the number of samples. As a consequence, the acquisition time will be reduced. The second problem discussed in this paper is represented by the accuracy. To preserve the informational consistency, the samples will be focused in the area of the spikes. For this, the algorithm identifies the extreme points of the amplitude frequency for the tested device and next, a spline interpolation is provided.

关键词： microwaves frequency amplitude sampling algorithm

来源：评论

学校读者我要写书评

暂无评论

Design and Application of High Fidelity sampling algorithm for Flight Test Data 5

Design and Application of High Fidelity Sampling Algorithm f...

引用

5th International Conference on Sensing, Measurement and Data Analytics in the Era of Artificial Intelligence, ICSMD 2024

作者： Zha, Xiaowen Deng, Guobao Feng, Can Tao, Liu Xue, Bowen Comac Flight Test Center Flight-Test Instrumentation Department Shanghai China

ISBN: (纸本)9798331529192

Flight test data is a critical element for characterizing aircraft performance and status, serving essential roles throughout stages such as design, manufacturing, flight testing, and operations. Generated by onboard data acquisition systems as strict time-series data, flight test data is utilized by engineers and data analysts for computation, analysis, and visualization to assess flight mission outcomes. Traditional client-server (C/S) architecture for data visualization involves reading all data points throughout the entire process and using graphics cards for sampling display. However, as the number of flight test data parameters grows exponentially and data volumes increase, data management applications have adopted distributed storage architectures. Consequently, data analysis and visualization have primarily shifted to browser-server (B/S) architecture modes. A single flight test parameter for one mission can generate tens of millions of data points, often requiring simultaneous display of multiple parameters. Due to browser performance limitations, reading all data points in realtime is impractical. Therefore, a high-fidelity sampling algorithm must be developed, tailored to the specific characteristics of flight test data. This algorithm should extract key feature points, adapt to display resolution, and ensure highfidelity visualization of flight test data, meeting the demands of B/S architecture analysis applications. © 2024 IEEE.

关键词： Data Analysis Flight Test Data High-Fidelity sampling algorithm Visualization Display

来源：评论

学校读者我要写书评

暂无评论

Landslide susceptibility mapping using an automatic sampling algorithm based on two level random sampling

引用

COMPUTERS & GEOSCIENCES 2019年 133卷 104329-000页

作者： Aktas, Hakan San, Bekir Taner Akdeniz Univ Dept Elect & Elect Engn Dumlupinar Blv TR-07058 Antalya Turkey Akdeniz Univ Dept Geol Engn Dumlupinar Blv TR-07058 Antalya Turkey

The aim of this study was to develop and use an algorithm to generate the automatic landslide susceptibility map. The proposed algorithm based on the two-level random sampling (2LRS) strategy and machine learning classification was generated using MATLAB. Performing automatic susceptibility mapping using machine learning classification requires an automatic and robust algorithm for the training of the constructed model. The proposed algorithm contains 20 steps and most of them have novel solutions for sampling. The user could also change the ratio of training and testing with this algorithm to perform automatic landslide susceptibility mapping. During the study, the 28 parameters used as input data sets which are the Digital Elevation Model (DEM), slope, aspect, plan curvature, profile curvature, the convergence index, the Topographic Wetness Index (TWI), the LS factor, the Normalized Difference Vegetation Index (NDVI), the Kaolinite Index, the Calcite Index, the OH Index, distance to fault lines, distance to channels, and 14 decorrelation stretched Advanced Spaceborne Thermal Emission and Reflection Radiometer (ASTER) bands. In this study, a total of 12 susceptibility models with different numbers of samples were automatically generated and tested using the produced algorithm in the study area, which contains active deep-seated rotational landslides (Alakir catchment area [Western Antalya, Turkey]). In addition, derivatives of the random forest (RF), support vector machines (SVM), and decision tree (DT) algorithms were compared in the MATLAB Classification Learner Toolbox according to their accuracies. The Medium Gaussian SVM has the highest level of accuracy (84%) among them. The constructed 12 models with different numbers of samples were also tested according to their spatial performance and their processing times. Then, the obtained area under curve values (AUCs) were obtained between 0.90 (in 360.009 s) and 0.84 (in 78.307 s).

关键词： Automatic landslide susceptibility SVM MATLAB Image processing sampling algorithm

来源：评论

学校读者我要写书评

暂无评论

Distributed Localization and sampling algorithm for 3-Dimensional Mobile Oceanic Sensor Networks

Distributed Localization and Sampling Algorithm for 3-Dimens...

引用

8th International Conference on Wireless Communications, Networking and Mobile Computing (WiCOM)

作者： Zhang Meiyan Cai Wenyu Zhejiang Water Conservancy & Hydropower Coll Dept Elect Engn Hangzhou 310018 Peoples R China Hangzhou Dianzi Univ Sch Elect Informat Hangzhou Peoples R China

ISBN: (纸本)9781612846835;9781612846828

Last several years have overseen a rapidly growing interest in Oceanic Sensor Networks (OSN in short), which consists of a variable number of sensor nodes working at underwater to monitor particular oceanic environment collaboratively. Compared with the traditional sensor networks, OSN has many additional challenges with acoustic communication environment such as large propagation delay, low communication bandwidth and high packet drop rate. Moreover, OSN belongs to one 3-Dimensional framework networks. Therefore, traditional localization schemes for 2-Dimensional terrestrial sensor networks should be transformed into 3-Dimensional localization problem. In this paper, we proposed a large-scale OSN framework for distributed localization first. And then a distributed range-based localization algorithm designed for 3-Dimensional OSN is illustrated in details. Localization accuracy and localized ratio are two metrics to verify our proposed scheme.

关键词： Oceanic Sensor Networks Localization Scheme Distributed algorithm sampling algorithm Mobile OSN

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：