microarray-classification

the classification of micro array gene expression samples using intelligent techniques The main objective of the work is to maximize the performance accuracy of the classifier so that the unknown microarray sample gets correctly classified to its respective class. The correct classification is of great concern to the medical researchers and biologists as the patient need to be treated accordingly. For efficient design of classifier, we first implement FFNN trained using Back Propagation (BP) Algorithm. Due to the disadvantage of BP Algorithm i.e. getting stuck in the local minima and long uncertain training process, we then extend to implement PNN. Finally, we compare the performance of FFNN and PNN by implementing SVM. The classifier showing the highest accuracy is judged by comparing the performance of FFNN, PNN and SVM.