In this paper, firstly a technique for confusion class recognition based on classification error distribution is proposed to recognize confusion class sets existing in the pre-defined taxonomy.
The classification error rate for normal and early stage DR samples reached 21.35% using a linear classifier and the leave-one-out method.
When training sets with uneven class sizes are used, the classification error based on C-Support Vector Machine is undesirably biased towards the class with fewer samples in the training set.
