搜索资源列表
lasso
- 使用lasso方法,对特征矩阵进行优化特征选择(向量选择),使其达到最优。-Use lasso method, optimize the characteristic matrix feature selection (vector selection), to the optimal.
F_Score
- 特征选择的一种方法,对数据矩阵进行f打分,然后降序排序,可以从中发现重要的数据。-A method of feature selection, to f score data matrix, and then descending order, you can find important data.
code
- 1采用遗传算法对男女生样本数据中的身高,体重,喜欢数学,喜欢文学,喜欢运动,喜欢模式识别共6个特征进行特征选择,并基于所得到的最佳特征采用SVM设计男女生分类器,并计算模型预测性能(包含SE,SP,ACC和AUC )。提示:可以用6位的0/1进行编码,适应度函数可以考虑类似 。-1 genetic algorithm for boys and girls in the sample data of height, weight, like math, like literature, like
Bat-Selection
- 这是一种蝙蝠特征选择,其目标是从一个给定的一组特征中找到最重要的信息,优化数据-This code use as optimization of data by row or coulmn Feature selection aims to find the most important information a given set of features
bGWO
- 这个文件夹包含实现二进制灰太狼优化器,以包装模式进行特征选择。 -This folder contains implementation for the binary grey wolf optimizer applied for feature selection in wrapper mode.
SVM-RFE-CBR-v1.3
- 在生物信息学中,SVM-RFE是一个强大的特征选择算法。这是一个不错的选择以避免过度拟合特性高的数量。-SVM-RFE is a powerful feature selection algorithm in bioinformatics. It is a good choice to avoid overfitting when the number of features is high.
SVMRFE.m
- 基于RFE特征选择方法的多分类特征排序,Matlab平台(Multi class feature ranking based on RFE method)
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
grayco
- 用3*3邻域计算每一个像素的灰度共生矩阵并计算对应的特征参数,最后再选择特定目标区域内的像素,计算在目标区域内的像素所对应的特征参数的均值和方差(Calculation of characteristic parameters of each pixel gray level co-occurrence matrix and calculate the corresponding 3*3 neighborhood, then select the specific pixels within t
ga_SVM_1
- 结合遗传算法和SVM,实现特征选择和SVM参数优化同时进行(Combining genetic algorithm and SVM, feature selection and SVM parameter optimization are carried out simultaneously)
svmcls
- 李荣陆老师做的文本分类器,特征选择方式包括全局和按类别选取,概率估算方法支持基于文档(布尔)统计和基于词频统计,支持三种特征加权方式,特征评估函数包括信息增益、互信息、期望交叉熵、X^2统计,文本证据权重,右半信息增益,分类方法包括支持向量机SVM和K近邻KNN,(text classifier that was written by Li Ronglu)
CNN
- 卷积神经网络的全部代码,已经调试通过了,可用于特征选择(Convolutional neural network all the code, has been debugged, and can be used for feature selection)
ypea106-real-coded-simulated-annealing
- 特征选择算法,可根据自己需要进行修改,简单易懂。(Feature selection algorithm can be modified according to its own needs, easy to understand.)
FastICA
- 独立成分分析matlab代码,进行特征降维与特征选择(ICA transform for feature selection)
MCFS_p
- 过滤式特征选择,测量特征之间相关性。仅限交流使用(correlation feature selection)
matlab 蚁群算法ACO_feature_selection
- 蚁群算法用与特征选择,针对传统蚁群聚类算法收敛速度过慢的问题,提出一种对蚁群算法进行改进的聚类算法。而数据的高维使数据具有稀疏、不可聚集等特性,使聚类算法实验效果精度低和耗时大,将邻域特征选择与聚类算法结合,提出了一种蚁群聚类优化的邻域特征选择算法(Ant colony algorithm and feature selection)
粒矩阵属性约简的启发式算法
- 基于矩阵的运算, 属性约简,特征选择,能够快速的找出最小约简属性子集(Boolean matrix attribute reduction Matrix based operations, attribute reduction and feature selection can quickly find the minimal reduct subsets)
UCI
- 里面含有连续型数据集,离散型数据集以及混合型数据集可以用于属性约简,特征选择等算法的实验仿真。以及直接导入weka软件。(It contains continuous data sets, discrete data sets and mixed data sets, and can be used for the experimental simulation of attribute reduction and feature selection algorithms. And import
svm_rfe
- 实现了SVM_RFE算法,进行特征选择并分类(Implementation of the SVM_RFE algorithm, feature selection and classification)
PCA_Monitoring
- PCA(principle component analysis)算法,可用于特征选择等,希望有帮助(principle component analysis)