I have a feature vector of dimension about 2000. All are histogram features.
I dont know which is effective in my case : applying SVM with RBF kernel or Chi-squared kernel ??
Could you please suggest me the effective kernel in my case?
Share
Sign Up to our social questions and Answers Engine to ask questions, answer people’s questions, and connect with other people.
Login to our social questions & Answers Engine to ask questions answer people’s questions & connect with other people.
Lost your password? Please enter your email address. You will receive a link and will create a new password via email.
Please briefly explain why you feel this question should be reported.
Please briefly explain why you feel this answer should be reported.
Please briefly explain why you feel this user should be reported.
I have a feature vector of dimension about 2000. All are histogram features.
I dont know which is effective in my case : applying SVM with RBF kernel or Chi-squared kernel ??
Could you please suggest me the effective kernel in my case?
In general Chi-squared and Intersection kernels are best suited for histogram style feature vectors. However, the results can often vary depending on the data, histogram normalization, kernel parameters etc. So the best way to choose the kernel type and the SVM and kernel parameters would be to do it empirically using cross-validation.
Also, FYI a svm is linear only if the kernel is linear. With chi-squared, Intersection, RBF and polynomial kernels, the svm will no longer be linear.