How to perform normalization after feature extraction process?

Question

Yuvaraj 2013-4-29

0
链接

此问题的直接链接

https://ww2.mathworks.cn/matlabcentral/answers/73910-how-to-perform-normalization-after-feature-extraction-process

hi After feature extraction, i got 2520*210 matrix. Now i want normalize these values separately by subject wise? Kindly help me???? Urgent...

1 个评论
显示 -1更早的评论隐藏 -1更早的评论

Walter Roberson 2013-9-5

http://www.mathworks.com/matlabcentral/answers/29922-why-your-question-is-not-urgent-or-an-emergency

此问题已关闭。

Answer 1

Image Analyst 2013-4-29

0
链接

此回答的直接链接

https://ww2.mathworks.cn/matlabcentral/answers/73910-how-to-perform-normalization-after-feature-extraction-process#answer_83782

I have no idea. Do you have 210 different features that you measured, and you have done that for 2520 subjects? What does normalization mean to you? Because it's not just some one single definition carved in stone that applies to every situation in the universe.

2 个评论
显示无隐藏无

Yuvaraj 2013-4-30

编辑：Image Analyst 2013-4-30

在 MATLAB Online 中打开

2520*210 is my total feature vector. I want to normalize all the data to 0-1. I don't know how to do loop ? by this formula i have to normalize the data.

s=xx-mean(xx)
norm = s/max(abs(s))

The program is:

[m n]=size(features) %2520*280
for i=1:m
    for j=1:n
        s(i,j)=(features (i)-mean(features(1:m)));
        %norm_features(i,:) = s(i,:)/ max( abs(s(i,:))); 
    end
end

Kindly help me.

Image Analyst 2013-4-30

Please answer my first question - you forgot to do that. I would think you'd need to normalize each feature (column or row) independently of all the other features because they're different features and most likely they don't have the same ranges.