Classification data pre-processing

Hai, I have a data set having two features. Each feature has 62 elements. These 62 elements are divided into 4 classes. But inter-class separation is not good. I have to use them for classification of data. How can i make them having good inter-class separation ?

for example : a=1,2,3,4,3,5,6,5,7,6,8,9,8,7,9 here first 5 elements belong to class-a, next 5 elements belongs to class-b, and last 5 elements belong to class-c. If the inter-class separation is not good, how can i make them having good inter-class separation ? Thanks in advance.