Hi @ANANDHI
To train a non-linear SVM classifier using a Gaussian kernel (also called RBF kernel) in MATLAB, the "fitcsvm" function can be used with the "KernelFunction" set to "rbf".
Assuming the features are stored in a matrix "features" and the corresponding class labels (e.g., 0 for benign and 1 for malignant) are stored in a vector "labels", the training can be done like this:
SVMModel = fitcsvm(features, labels, 'KernelFunction', 'rbf');
Each row in "features" should represent one sample (tumor), and each column should represent one of the three extracted features. The "labels" vector should have the same number of elements as the number of rows in "features".
This will train an SVM with a Gaussian (RBF) kernel using the three features provided.
Below is the MathWorks documentation link to know more about "fitcsvm" :
Hope this helps!
