how can I constrain weights of linear classifier in MATLAB?

I am graduate student studying system neuroscience in Korea. I have a problem and need your help... problem is like below.

There are a number of neurons and we know their firing rates on -3s, -2s, -1s, like

neuron index :

                  1     2     3     4

firing rata

         0s:      1     1    -1     1
        -1s:     -1    -1     0    -1
        -2s:      0     0     0     0
        -3s:      1     1     1     1

By using these fire rate history, I want to predict current(0s) firing rate of neurons whether it is 1 or -1. I am using linear SVM and varying weights of each time point. But, I want to constrain absolute value of weight of recent firing rate is always larger than that of past like

weight of

   -1s:  -0.8

   -2s:  0.3

   -3s:  -0.1

How can I realize this idea on linear classifier like linear SVM on MATLAB? (of course, other linear classifier would be welcomed)

please help me...

Answer 1

[Rakesh Chavan]

Hi,

It is possible to create a linear SVM model using the 'fitcsvm' function based on the training data. http://www.mathworks.com/help/stats/fitcsvm.html

You can use the 'predict' function to for classifying new data using a trained SVM classifier: http://www.mathworks.com/help/stats/compactclassificationsvm.predict.html

A general overview of the process is given in the following link: http://www.mathworks.com/help/stats/support-vector-machines-svm.html

For generating a linear discriminant analysis classifier the 'fitcdiscr' can be used. The weights option for this function can be used to specify custom weights. Kindly refer to the following link: http://www.mathworks.com/help/stats/fitcdiscr.html