MATLAB: Is SVM resilient to noise

machine learningpcaStatistics and Machine Learning Toolboxsvm

I have tranning set composed of 36 features. when I calculated "explained" value of PCA using Matlab. I notice that only the first 24 components are important.
My question is, would I gain a better accuracy (prediction) if I omit the reset of the components (the other 12 components). Or SVM is very resilient to noise which means that regardless whether I removed the other 12 components or not. performance will not change that much.

Best Answer

I would think it would matter where the noise was. If it's far away from the dividing line, then it doesn't make any difference. If it's close to the dividing line, then yeah, it makes a huge difference.
Related Question