基于模式分类的双支持向量机（Matlab代码实现）

💥💥💥💞💞💞欢迎来到本博客❤️❤️❤️💥💥💥

🏆博主优势：🌞🌞🌞博客内容尽量做到思维缜密，逻辑清晰，为了方便读者。

⛳️座右铭：行百里者，半于九十。 

目录

💥1 概述

📚2 运行结果

🎉3 参考文献

🌈4 Matlab代码实现

文献来源：

💥1 概述

一种二进制SVM分类器，通过解决两个相关的SVM类型问题来确定两个非平行平面，每个问题都比传统SVM中的小。Twin SVM公式是通过广义特征值实现近端SVM的精神。在几个基准数据集上，Twin SVM不仅速度快，而且具有良好的泛化能力。双SVM对于自动发现数据的二维投影也很有用。

📚2 运行结果

clc;clearvars;close all;rng default;​% Read Datarun('data_heartstatlog.m');​% Get Data and Labelsfeatures=data(:,1:end-1);labels=data(:,end);​% Normalize labelslabels(labels==2)=-1;​​% Separate training and test data (80:20 split)total_samples=size(features,1);train_samples=round(0.8*total_samples);​% Define training and test samplesxTrain=features(1:train_samples,:);yTrain=labels(1:train_samples,:);xTest=features(train_samples+1:end,:);yTest=labels(train_samples+1:end,:);​% Define hyperparameter valuesC1=0.1; C2=0.05;​% Run Twin SVM (Linear) [ yPred, accuracy ] = LinearTWSVM( xTrain, yTrain, xTest, yTest, C1, C2 );  disp('Accuracy (Linear) is'); accuracy  % Run Twin SVM (Kernel)kernel_type=3; kernel_param=0.95;​% Run Kernel Twin SVM[ yPred, accuracy ] = KernelTWSVM( xTrain, yTrain, xTest, yTest, C1, C2, kernel_type, kernel_param ); disp('Accuracy (Kernel) is');accuracy 

Heart statlog data in useAccuracy (Linear) is​accuracy =​   83.3333​Accuracy (Kernel) is​accuracy =​   75.9259​>>

clc;clearvars;close all;
rng default;

% Read Data
run('data_heartstatlog.m');

% Get Data and Labels
features=data(:,1:end-1);
labels=data(:,end);

% Normalize labels
labels(labels==2)=-1;

% Separate training and test data (80:20 split)
total_samples=size(features,1);
train_samples=round(0.8*total_samples);

% Define training and test samples
xTrain=features(1:train_samples,:);
yTrain=labels(1:train_samples,:);
xTest=features(train_samples+1:end,:);
yTest=labels(train_samples+1:end,:);

% Define hyperparameter values
C1=0.1; C2=0.05;

% Run Twin SVM (Linear)
 [ yPred, accuracy ] = LinearTWSVM( xTrain, yTrain, xTest, yTest, C1, C2 );
 
 disp('Accuracy (Linear) is');
 accuracy
 
 
% Run Twin SVM (Kernel)
kernel_type=3; kernel_param=0.95;

% Run Kernel Twin SVM
[ yPred, accuracy ] = KernelTWSVM( xTrain, yTrain, xTest, yTest, C1, C2, kernel_type, kernel_param );
 
disp('Accuracy (Kernel) is');
accuracy

Heart statlog data in use
Accuracy (Linear) is

accuracy =

   83.3333

Accuracy (Kernel) is

accuracy =

   75.9259

>> 

🎉3 参考文献

[1]Jayadeva, Khemchandani, R., and Suresh Chandra. "Twin support vector machines for pattern classification." IEEE Transactions on pattern analysis and machine intelligence 29.5 (2007): 905-910.

🌈4 Matlab代码实现