Classify maintenance request is one of the processes in the large software system to support maintainers in doing their daily maintenance tasks more effectively. Categorizing these maintenance requests are an essential requirement in managing the maintenance request for software maintainer and need a great effort as well as determining classification. Hence, this paper presents the framework from the use of three different classification approaches, namely Bayesian model Decision Tree and Logistic regression. We show that nai??ve Bayesian classifier, Decision Tree and Logistic regression can be used to accurately classify issues into maintenance type.