With the development of web technology, researchers had started using blog data in many research aspects and twitter messages is one of them. However, these data are un-organized and thus, it should be organized before gather the information. Classification is one way of organizing the twitter messages. SVM and Naive Bayes classifiers are the most popular classification methods which are often use for text classification. Theoretically, it proves that Naive Bayes performs more faster than any other classifiers with less error. However, this depends on how the situation achieves the Naive Bayes assumptions as naive Bayes assumes that the features are independent. This paper presents a practical experiment to choose a high perform classification method and the theoretical reasons for the high performed classification.