A Hybrid Classification Based on Machine Learning Classifiers to Predict Smart Indonesia Program

Data Mining applies mining techniques to learning-related data. Predicting students who are eligible for the Smart Indonesia Program (PIP) is complicated because several important components can be taken into consideration in deciding whether or not the student deserves. To help this, a survey was conducted to obtain a clear view. This survey was conducted with the help of accuracy in predicting students who were eligible for the PIP. There are two factors involved in this selected process, attributes of predictor and prediction methods. The core purpose of this study is to predict students who are eligible for the Smart Indonesia Program (PIP) using the mining method idea. This study focused on the study of a hybrid algorithm and compared with machine learning classifiers such us Iterative Dichotomiser 3 (ID3), artificial neural network, Naïve Bayes, and K-nearest neighbor (KNN). The study discusses a hybrid classification based on several machines learning classifiers to predict students who are eligible for the Smart Indonesia Program (PIP). In this study, we used four measurements to analyze the performance of the proposed algorithm. The findings of the research are the hybrid of classification able to predict the feasible receiver of PIP with accuracy 89.38 percent and the best accuracy is the ANN method. In deep, the system of a hybrid algorithm is better and consistent than other classification algorithms with F1 measure 94 percent.