助手标题  
全文文献 工具书 数字 学术定义 翻译助手 学术趋势 更多
查询帮助
意见反馈
   feature selection 的翻译结果: 查询用时:0.011秒
图标索引 在分类学科中查询
所有学科
计算机软件及计算机应用
自动化技术
互联网技术
电信技术
武器工业与军事技术
数学
电力工业
工业通用技术及设备
机械工业
更多类别查询

图标索引 历史查询
 

feature selection     
相关语句
  特征选择
     Feature Selection Based on CHI and Genetic Algorithm
     基于CHI与遗传算法的特征选择
短句来源
     New Fuzzy Feature Selection Algorithm: Ⅱ
     模糊特征选择新算法:Ⅱ
短句来源
     5. Feature selection.
     5.特征选择
短句来源
     The method of recognition and feature selection and extraction are the core of recognition.
     在这一识别问题中,识别方法及与之相应的特征选择和抽取是问题的核心。
     In this paper, a feature selection method from the point of the characteristic of spatial data, named MEFS (maximum entropy feature selection), is proposed.
     从空间数据本身的特性出发,提出一种特征选择方法MEFS(maximum entropy feature selection).
短句来源
更多       
  特征提取
     Key Feature Selection for NIDS Using GA and SVM
     利用GA与SVM对NIDS进行关键特征提取
短句来源
     Feature Selection of DGA Data Based on Transformer Fault Classification
     基于变压器故障分类的DGA特征提取
短句来源
     (3) We employed SVM and the features of vehicle profile, acoustic and seismic signals to recognize the types of vehicles, and analyzed the effect of different feature selection and extraction methods on the classification accuracy.
     (3)分别利用车辆轮廓特征、声音信号和地表震动信号,结合支持向量机分类原理,对车辆类型进行了分类研究。 分析了不同特征提取和特征选择方法对分类准确率的影响。
短句来源
     Feature Extraction and Feature Selection Based on Wavelet and Genetic Algorithm
     基于小波与遗传算法的特征提取与特征选择(英文)
短句来源
     A Study of Feature Selection Method Based on Support Vector Machine and Its Application
     基于支持向量机的特征提取方法研究与应用
短句来源
更多       
  特征选取
     OFFSS (Optimal Fuzzy-valued Feature Subset Selection) is a new fuzzy-valued feature selection method that selects an optimal feature subset from the feature space by considering both the overall overlapping degree between two classes of examples.
     OFFSS (Optimal Fuzzy-valued Feature Subset Selection)是一种新的模糊值特征选取的方法,是基于两类事例集合的重叠程度来选取特征空间中最优特征子集。
短句来源
     Feature selection is NP-Hard problem.
     特征选取是一个NP-Hard问题。
短句来源
     This paper reports a comparative study with CHI ,IG,DF and MI feature selection methods for Chinese Web pages.
     针对中文网页,比较研究了CHI、IG、DF以及MI特征选取方法。
短句来源
     DNA Microarray Data Analysis Based on Optimal Orthogonal Centroid Feature Selection
     基于最优正交质心特征选取的DNA微阵列数据分析
短句来源
     Decision Rules Mining Method by Integrating Feature Selection and Discretization
     与特征选取和离散化集成的决策规则挖掘方法
短句来源
更多       
  属性选择
     Firstly, making a comprehensive survey of feature selection for unsupervised learning in the past research, these are theoretical foundations of my paper. Secondly, we introduce a novel methodology ULAC (Feature Selection for Unsupervised Learning Based on Attribute Correlation Analysis and Clustering Algorithm).
     在已经深入了解和体会现有发展的基础上,提出一种新型的属性选择模型——无指导学习环境下基于属性相关性分析和聚类算法的属性选择方法ULAC(Feature Selection for Unsupervised Learning Based on Attribute-Correlation Analysis and Clustering Algorithm)。
短句来源
     The essence of rough set theory is data reduction which can be used for feature selection has already been applied to some algorithms successfully.
     粗糙集合理论的精髓是数据约简,利用数据约简可以处理属性选择问题,目前已有一些属性选择算法的研究开始关注于应用粗糙集合理论,并初步得到实验验证.
短句来源
     A TWO-PHASE METHOD OF NEURAL NETWORK FEATURE SELECTION
     一种两阶段的神经网络属性选择方法
短句来源
     Feature selection was considered as feature selection in supervised learning from traditional view.
     过去传统意义上的属性选择通常是指在有指导学习环境下的属性选择
短句来源
     This paper presents a method of neural networks feature selection based on data attributes importance ranking.
     提出一种基于数据属性重要性排序的神经网络属性选择方法。
短句来源
更多       

 

查询“feature selection”译词为其他词的双语例句

 

查询“feature selection”译词为用户自定义的双语例句

    我想查看译文中含有:的双语例句
例句
为了更好的帮助您理解掌握查询词或其译词在地道英语中的实际用法,我们为您准备了出自英文原文的大量英语例句,供您参考。
  feature selection
The testing and experimental results of feature selection show that NQGA presents good search capability, rapid convergence, short computing time, and ability to avoid premature convergence effectively.
      
In addition, the machinery fault data were analyzed by this method, and the attribute reduction sets were obtained further to satisfy the demand of feature selection in machinery diagnosis.
      
Then, the parameters for feature selection and kernel selection are learned simultaneously by maximum a posteriori for given samples and uncertain labels.
      
The key points are feature selection and classification.
      
Evaluation on this system proved that the feature selection has important impact on the system performance, and the post-processing explored has an important contribution on system performance to achieve better results.
      
更多          


This paper describes the algorithms and personal experience of applying statisticalpattern recognition techniques to seven spectral images, which were taken by an aero-plane from a medium altitude. The subject is discussed in the following three sections:(1) For multicategory classification, feature selection through divergency maximization is more appropriate than K-L transform. But, for the mixture of ratio and original images, the divergence formula for normal populations is not applicable.(2) Spatial...

This paper describes the algorithms and personal experience of applying statisticalpattern recognition techniques to seven spectral images, which were taken by an aero-plane from a medium altitude. The subject is discussed in the following three sections:(1) For multicategory classification, feature selection through divergency maximization is more appropriate than K-L transform. But, for the mixture of ratio and original images, the divergence formula for normal populations is not applicable.(2) Spatial smoothing before classification will improve the results. The selected smooth-ing algorithm, however, should not affect the multi-spectral features excessively.(3) The proposed classification algorithm, whieh combines the maximum likelihoodand table lookup algorithm, will speed up the processing by about a factor oftwo, while the results are the same as by using the maximum likelihood algorithmonly.

本文介绍作者在对中高度飞机所摄7谱图象进行统计模式识别时所用的一些方法和体会。(1)在多类情况下,用分散度作准则选择特征较K-L变换为宜。但对于比例图和原始图的混合,正态情况下的分散度公式并不适用。(2)在分类前先作空间平滑处理可改进分类结果。选用平滑算法时应注意不要破坏原有的多谱特性。(3)把查表法和最大似然法相结合可提高速度1倍左右而结果与最大似然法相同。

This paper describes the principles and implementation of scene matching used in position location systems in brief. Several essential problems in scene matching technique are reviewedt measures of similarity (MAD, MSD, PROD, NPROD, Pair Functions etc. )are outlined; fast search me-thods(resolution from coarse to fine, quantization from coarse to fine, feature extraction, modified SSDA etc. )are introduced; probability of acquisition for max. or min. algorithm is given; preprocessing of images (filtering, restoration,...

This paper describes the principles and implementation of scene matching used in position location systems in brief. Several essential problems in scene matching technique are reviewedt measures of similarity (MAD, MSD, PROD, NPROD, Pair Functions etc. )are outlined; fast search me-thods(resolution from coarse to fine, quantization from coarse to fine, feature extraction, modified SSDA etc. )are introduced; probability of acquisition for max. or min. algorithm is given; preprocessing of images (filtering, restoration, whitening etc. )is considered; geometrical distortion(sync-hronization error, rotation, scale factor, perspective distortion)and its effects are analyzed, and the methods for overcoming such effects are presented.After then, two noticeable important techniques for homing seeker, the feature matching and the correlation tracking, are remarked. Several methods of feature extraction are introduced for such features in scene matching as invariant moments, edge features, line features, planar features, vertex features etc. . The emphasis is put on feature selection, update and replacement in terminal target homing.Finally, the block diagrams of a combined optimum aided navigation system and a homing seeker using correlation tracking are given.

本文概述了匹配定位原理、方法及相似度度量、快速搜索、正确匹配概率、预处理、几何失真及其改善等问题,还指出了特征匹配及匹配寻的跟踪等两个值得重视的发展趋势。最后给出了相关匹配技术在组合导航及导引头中应用的结构组成。

In this paper we discuss the feature selection in screening cervical cells by meang of optical power spectra. After analyzing morphological characteristics of normal cells and cancer cells and features of corresponding spectra, we have found the key for recognition. This feature alone can give a correct-recongition rate up to 94.5%, higher than the rate of 92% obtained before by using six features in a similar experiment[1]. We also present the relation between the correct-recognition...

In this paper we discuss the feature selection in screening cervical cells by meang of optical power spectra. After analyzing morphological characteristics of normal cells and cancer cells and features of corresponding spectra, we have found the key for recognition. This feature alone can give a correct-recongition rate up to 94.5%, higher than the rate of 92% obtained before by using six features in a similar experiment[1]. We also present the relation between the correct-recognition rate and the dimension of feature space. The results indicate a fall-down of the correct-rate when the feature number is increased improperly.

本文研究了利用光学功率谱识别宫颈细胞工作中的特征选择问题。从正常细胞与癌细胞的形态特征以及相应的频谱分布的规律性出发,找出了识别分类的最佳特征。当采用此单个特征时,正确识别率可达94.5%,比以前类似工作中用6个特征所达到的识别率92%有所提高。若采用3个特征,则正确识别率可达95.5%。此外,本文还给出了分类正确率与特征维数的关系曲线。结果表明,当采用更多数目的特征时,反而会使分类正确率下降。

 
<< 更多相关文摘    
图标索引 相关查询

 


 
CNKI小工具
在英文学术搜索中查有关feature selection的内容
在知识搜索中查有关feature selection的内容
在数字搜索中查有关feature selection的内容
在概念知识元中查有关feature selection的内容
在学术趋势中查有关feature selection的内容
 
 

CNKI主页设CNKI翻译助手为主页 | 收藏CNKI翻译助手 | 广告服务 | 英文学术搜索
版权图标  2008 CNKI-中国知网
京ICP证040431号 互联网出版许可证 新出网证(京)字008号
北京市公安局海淀分局 备案号:110 1081725
版权图标 2008中国知网(cnki) 中国学术期刊(光盘版)电子杂志社