Comparative Study of Back Propagation Neural Network and Support Vector Machine to Classify Hypertension Gene Sequences from Relative Codon Frequency

In this work, relative codon frequency has been considered to specify gene sequences. Implementing BPNN and SVM hypertension gene sequences have been categorized from others. A single gene sequence has been converted into 19 dimensional feature vector and then the Back Propagation Neural Network and Support Vector Machine have been trained using the feature vectors of all gene sequences. Since relative frequencies are not dependent on lengths, sequences with various lengths are used here. Accuracy rates of Support Vector Machine and Back Propagation Neural Network have been compared using different no. of samples at the training and testing phase. In this experiment, the accuracy rate has been increased proportionally with the size of samples. It is found that in most of the cases, BPNN is better than SVM.