first post here.
I'm pretty new to machine learning and am currently getting to grips with SVMs.
I currently have a set of data in a Numpy array and I am struggling to get any decent sort of accuracy. I understand that there is a way to tune the parameters of an SVM to make it more accurate for the data which I have.
A sample of the data I have is shown below:
A B C D E F G H I J Class
62 0 1 12 231 0 5 7 0 1 0
17 0 2 412 754 1 4 8 1 2 0
54 1 0 104 100 0 4 2 1 2 1
18 0 2 189 172 1 2 8 0 2 1
41 1 0 311 612 1 1 3 0 1 0
This is just a sample, I have over 7000 records.
I currently have the data in (apart from class) in a Numpy array named X, and the class data in a Numpy array named y.
I've tried feature selection but the highest accuracy i'm getting is 85% and I'd like to know how to be able to tune the parameters of the SVC to make it more accurate for my data.
Any help here would be a big help.
Thanks in advance, Isaac.
