Classification and prediction of student's educational status using data mining techniques

Data mining and discovering hidden knowledge and patterns from data in educational systems enables the decision makers in higher education domain to improve educational processes such as planning, enrollment, assessment and counseling. The goal of this study is to classify students and prediction of their educational status using data mining techniques. In this research, we have tried to build proper models for predicting students’ educational status in the following semester using appropriate data preparation of their demographic and educational history and applying it to different classification techniques including decision tree, logistic regression, k-nearest neighbor and neural networks. Finally we have compared the results obtained from these techniques and offered k-nearest neighbor and neural networks as the best models for students’ classification. On this basis, proposed models can be utilized as a decision supporting tool in educational systems.