搜索资源列表
SVMpreprocess
- 实现了文档预处理,特征选择,特征权值计算,用于SVM分类器-To achieve a document preprocessing, feature selection, feature weight calculation for the SVM classifier
kfcfk
- 特征的选择与提取 模式识别 统计分量 排序-Feature selection and extraction of statistical pattern recognition component to sort
pattern-recognition-simulation
- 用mushrooms数据对模式识别课程讲述的各种模式分类方法[线性分类,Bayesian分类,Parzen窗,KNN]和特征选择和降维方法[PCA,LDA]进行了模拟,并给出了各类分类方法的结果,-It s the simulations about linear classification ,Bayesian ,Parzen and KNN of pattern recognition .And ,It gives the results.
mboxplot
- 自己在进行分类时写的用于多变量分布浏览,特征选择的函数,可快速查看各特征分布情况,便于后续特征选择-use to display character value
face
- 完整的表情识别系统一般包括人脸表情图像捕获、预处理、人脸检测与定位、 人脸分割与归一化、人脸表情特征提取、人脸表情识别。本文着重研究了人脸表 情特征提取、特征选择及表情分类等关键问题,并提出了一些改进的方法,同时 进行了仿真实验-Complete expression recognition systems typically include facial expression image capture, preprocessing, face detection and loca
pcaica
- pca与ica相结合的特征选择,进行主成分分析以后,再对所得特征进行独立成分分析-the combination of pca and ica for feature selection,after Principal component analysis, the resulting characteristics is has a independent component analysis
dataset_603551
- 文件为为本分类器,可以实现切词、特征选择、权重计算等功能-File as classifier, can be achieved segmentation, feature selection, the weight calculation function
DF
- 文件为DF特征选择的代码实现,用java编写,欢迎大家下载-File DF feature selection code, written in java, welcome to download
Single-Pass
- 改进Single-Pass聚类算法,包括分词、tfidf计算、卡方检验特征选择-Improved Single-Pass Clustering Algorithm
Relief
- 利用距离求解的特征选择算法,针对两类问题的Relief算法-feature selection
BaB
- 特征选择,提取的一种算法,B and B算法的简单描述-An algorithm of feature selection and extraction
Bpes
- Bpes是基于特征提取和特征选择的算法,多用于已分类的数据集中的边界数据的检测,对于边界数据中的未分类数据及模糊数据具有很好的检测效果。可适用于分类器的构建。-Bpes is based on the feature extraction and feature selection algorithms, used to have more classification, the data of the boundary of the data set of unclassified data
gmdh_feature_select
- 本程序是一个基于matlab的特征选择程序,可以用于预测方面,实现效果不错-This procedure is a feature selection based on matlab program that can be used to predict aspects of implementation effect is good
pluslr
- 使用了MATLAB对信号进行特征选择,使用的信号有图像信号,一维信号等。对学习信号特征有参考价值。-Using the MATLAB signal for feature selection, use of image signal of the signals, and the one dimensional signal, etc.Characteristics of learning signal has a reference value.
光流法运动估计OpticalFlow
- 利用openCV,首先得到图像中的强边界作为跟踪的特征点,调用函数,输入两幅连续的图像,并在第一幅图像里选择一组特征点,输出为这组点在下一幅图像中的位置。再把结果过滤,去掉不好的特征点。把特征点的跟踪路径标示出来。(Using openCV, we use goodFeaturesToTrack function to get strong edges in the image as the feature point tracking, next to call calcOpticalFlow
基于极限学习G-score
- G-score是一个特征排序的准则,极限学习机结合G-score是一种filter+wrapper的混合特征选择算法(G-score is a criterion of feature sorting. Limit learning machine combined with G-score is a hybrid feature selection algorithm of filter+wrapper)
降维与特征选择
- 在machine learning中,特征降维和特征选择是两个常见的概念,在应用machine learning来解决问题的论文中经常会出现。 对于这两个概念,很多初学者可能不是很清楚他们的区别。很多人都以为特征降维和特征选择的目的都是使数据的维数降低,所以以为它们是一样的,曾经我也这么以为,这个概念上的误区也就导致了我后面对问题的认识不够深入。后来得到老师的指点才彻底搞清楚了两者的关系,现总结出来与大家分享。(Feature reduction and feature sele
REC-FSA-master
- 利用信息熵聚类对故障多特征量进行特征选择(Feature selection by using information entropy clustering for multiple features of fault)
FSASL-master
- 该方法通过计算核空间距离从而来对样本进行特征选择。(The method is used to select the features of the samples by calculating the distance of the nuclear space.)
信息论特征选择KDD Code
- 基于信息熵的特征选择算法,评价每个属性与分类的关联信息,评价属性,进行特征选择(Feature selection algorithm based on Information Entropy)