✅作者简介：热爱科研的Matlab仿真开发者，修心和技术同步精进，matlab项目合作可私信。
🍎个人主页：Matlab科研工作室
🍊个人信条：格物致知。
⛄ 内容介绍
支持向量机(英文简称SVM)可以很好地应用在数据分类及预测上,由于SVM在数据挖掘中具有通用性好,有效性,计算简单,理论完善等优点,所以得到了广泛的应用,文章利用matlab软件,基于SVM实现了对意大利葡萄酒的分类和预测.
⛄ 部分代码
clc
close all
addpath(genpath(pwd))
load banana
%
label = data(:,3);
data = data (:, 1:2);
% parameter setting
c = 0.8;
g = 0.05;
% train svm model
cmd = ['-s 0 -t 2 ', '-c ', num2str(c), ' -g ', num2str(g), ' -q'];
model = libsvmtrain(label, data, cmd);
[~, acc, ~] = libsvmpredict(label, data, model);
% meshgrid
d = 0.02;
[X1, X2] = meshgrid(min(data(:,1)):d:max(data(:,1)), min(data(:,2)):d:max(data(:,2)));
X_grid = [X1(:), X2(:)];
% set grid point label (only as input parameter)
grid_label = ones(size(X_grid, 1), 1);
% predict grid point labels
[pre_label, ~, ~] = libsvmpredict(grid_label, X_grid, model);
%
figure
% set(gcf,'position',[300 150 420 360])
color_p = [150, 138, 191; 220, 94, 75]/255;
color_b = [218, 216, 232; 244, 195, 171]/255;
hold on
gscatter(X_grid (:,1), X_grid (:,2), pre_label, color_b);
legend('off')
axis tight
%
ax(3:4) = gscatter(data(:,1), data(:,2), label);
set(ax(3), 'Marker','o', 'MarkerSize', 6, 'MarkerEdgeColor','k', 'MarkerFaceColor', color_p(1,:));
set(ax(4), 'Marker','o', 'MarkerSize', 6, 'MarkerEdgeColor','k', 'MarkerFaceColor', color_p(2,:));
legend('off')
set(gca, 'linewidth', 1.1)
title('Decision boundary')
axis tight
⛄ 运行结果
⛄ 参考文献
[1]付略, 周少华, 彭勃,等. 基于最小二乘支持向量机算法的南宋官窑出土瓷片分类[J]. 硅酸盐学报, 2008, 36(008):1183-1186.
