Classifiers such as KNN, SVM, Artificial Neural Network (ANN), Probabilistic Neural Network (PNN), Hidden Markov Model (HMM), etc. are used for various applications in real time. Each of the classification schemes previously mentioned has its own unique properties and associated strengths and problems. In KNN, the major limitation is that it uses all features in distance computationally severe, chiefly when the size of training set grows. Beside this, the accuracy of k-nearest neighbor classification is severely degraded by the presence of noisy or unrelated features, particularly when the number of attributes grows. In PNN, limitations is that it is slower than multilayer perceptron networks at classifying new cases and it requires more memory space to store the model. ANN, it performs better than other than other classification method with high dimensional features and contradictory data. But the high computing cost which consumes high CPU and physical memory usage is the main disadvantage of ANN. Bayesian approach outstanding with its simplicity and low computational cost in both the training and classifying stage and it has been widely implemented in various types of domains and applications. But this generative method has been reported to be less accurate than the discriminative methods such as SVM. SVM has shown to be more accurate than other classification approaches [14].