Prediction of student’s performance through educational data mining techniques

Many educators have worried about the failures of students through academic education. Thus, a variety predictions been applied to general information including culture, social, and economic which wasn’t related student performance. We gathered an actual dataset from three years stages Mustansiriyah University in Iraq. The consists without any socioeconomic data, it includes forty-four undergraduate with thirteen attributes. proposed model that explains correlation between two main subjects are, mathematics, control systems. This study aimed identify failure systems subject third year depending on features mathematics first second years. Three algorithms were Naïve Bayes, support vector machine, multilayer perceptron. Since was imbalanced, this leads appear overfitting problem results so synthetic minority oversampling technique utilized solve problem. Our show machine algorithm proves efficient classification after technique. accuracy classifiers measured confusion matrix using Waikato environment for knowledge analysis (WEKA) tool its metrics.