特别是支持向量机的特征选取和信赖度估计方法。
In particular methods for confidence estimation and feature selection with Support Vector Machines will be described.
第二，介绍了一种基于法矢量权重的特征选取方法。
Secondly, introduce a feature selection method based on the weight of normal from SVM model.
应用特征选取和模式聚合理论以降低特征空间维数。
This paper employs feature selection theory and pattern aggregation theory to reduce feature space dimension.
本文提出一种基于属性重要度的启发式特征选取算法。
In this paper, we introduce an algorithm which is using the importance of attribute with heuristics for feature selection.
分析了常用的几种特征选取方法，提出了改进互信息算法。
Several feature selection methods were analyzed. An improved mutual information algorithm was proposed.
使用普通英文文本和中文网页评测特征选取方法的结果是一致的。
The experimental results are similar using plain English text and Chinese Web pages to evaluate feature selection methods.
分析了产生差异的原因，并提出一种适合于中文环境的特征选取方法：组合特征选取方法。
The reason of difference is analyzed and a new method named Combined Feature Selection is put forward.
目前已有不少特征选取方法应用于基因表达数据，比如顺序前进法、遗传算法、信噪比指标等。
Now there are several feature selection methods applied to gene expression data, such as Sequential Forward Selection, GA, S2N and so on.
分析了常用的几种特征选取方法，提出了改进互信息算法。实验结果显示改进的互信息算法是可行的。
Several feature selection methods were analyzed. An improved mutual information algorithm was proposed. The experiment results showed that improved mutual information algorithm was feasible.
在此基础上提出了一种基于互信息特征选取的改进算法，该算法加强了互信息为负值的特征在分类中的作用。
So a new method based on improved mutual information (IMI) was presented, which strengthened the mutual information through reversing those features whose mutual information was negative.
这样，结合文献的研究成果，即可依据不同驱动类型和渗流特征选取参数，对油田产量进行全过程拟合和预测。
In this way, the whole process the field production will be matched and predicted by using selecting parameters representing different drive types and combining the study results of literature.
该方法将CHI值特征选取和粗糙集理论充分结合，避免了用粗糙集对大规模决策表进行特征约简，同时避免了决策表的离散化。
The method combined CHI value feature selection and rough set theory fully so as to avoid both feature reduction on a large scale decision table and the discretization of the decision table.
提出了一种阈值选取与边缘检测相结合提取红外图像特征的方法。
An algorithm using threshold selection and edge detection to extract target features from infrared image was proposed.
实验结果显示选取适当的特征数目、使用好的分词技术、使用命名实体识别技术都能改进中文话题追踪系统的性能。
The experimental result shows that the tracking performance will be improved by selecting proper features, using word segmentation system and using named entity recognition system.
然后利用直方图法对每一特征作了特征选择，如此共选取了21个特征。
Then trait selection has been for every trait using the histogram method and 21 traits in all have been selected.
本章选取了“在路上”电影叙事结构上的一些重要特征，以此来发掘此类影片在故事、结构与思想之间的贯通性。
Through analyzing some important narrative characteristics of on the road movie, this part tries to unearth the connection among the story, structure and theory of such kind of movies.
牌照字符识别是车牌识别系统中关键的一步，而字符识别的关键在于有效特征的选取。
License plate character recognition is a main step in license plate recognition system, but its emphasis is to choose efficient features.
为了使图像跟踪系统能根据图像特征自适应选取分割算法，给出了一种基于粗糙集理论的图像分割智能决策方法。
In order that the image tracking system can select segmentation algorithm itself according to the object image feature, this paper presents an intelligent decision method of image segmentation.
特征参数的选取对整个语音识别系统的实时性、鲁棒性有很大的影响。
The chosen speech feature parameters have great effects on robustness and real-time of the speech recognition system.
在此基础上主要就镜头分割、关键帧选取、音频特征分类以及视频中文字区域的检测和分割提出了相应改进的算法。
Further it mainly aims at the segment of shot, selection of key frame, classify of audio characteristic and detection of caption-region in video and presents corresponding improved approach.
利用历史资料实现了海域水深注记点的自动内插、海底地形特征点的自动选取和等深线的自动生成。
Automatic interpolation of charted depth annotated point, automatic selection of Marine topography feature point and automatic build of bathymetric contour come true by using history data.
通过适当的选取图像特征，实现了摄像机工作空间运动目标跟踪的视觉伺服任务，并采用扩展卡尔曼滤波控制方法完成机器人视觉伺服控制。
Floating object tracking in the whole working space of a camera can be obtained through selecting image features, and Kalman filtering control was adopted to realize visual servo control of robot.
基于支承法兰盘的结构特征，通过目标函数的建立、设计变量的选取以及约束条件的确定建立了优化设计的有限元模型，并对法兰盘进行了静力分析。
Based on the structural characteristics of bearing flange, the optimization model is established according to the objective function, design variables and constraints, and make a static analysis.
目的：采用粗糙集方法选取相关特征色谱峰用于黄芪药材的产地鉴别。
Objective: Correlative characteristic peaks of HPLC selected by rough sets were employed to identify the habitats of Radix Astragali.
对集装箱检测系统探测器信号的统计特征进行分析，选取合适的图像扫描参数对提高检测图像质量有十分重要的意义。
It is very important to analyze the statistic character of detector signal of container inspection system and further to choose suitable image scanning parameter.
提出了依据系统输出分布的特征变量选取方法，并提出了依据该项信息的建模和控制方法。
The method of modelling and control based on characteristic variables chosen according to system output distributions is presented.
论文讨论了如何选取声学特征来表征不同发音者的语音个人特性。
The paper deals with the selection of acoustic phonetics of the speaker characteristics.
基因表达数据的特征基因选取和肿瘤样本分类问题是基因微阵列技术的挑战性课题之一。
The problem of feature gene selection and tumor samples classification of microarray gene expression data is one of challenges of gene microarray technology.
基因表达数据的特征基因选取和肿瘤样本分类问题是基因微阵列技术的挑战性课题之一。
The problem of feature gene selection and tumor samples classification of microarray gene expression data is one of challenges of gene microarray technology.
应用推荐