Abstract
Education is the main foundation in shaping the future of students. In this era of technological development and global competition, it is important for schools to provide quality education, and also help students prepare for their next steps after graduating from school. The question of whether students will continue to higher education or immediately enter the world of work is a very important one. The research objects are Lagger Score, Family Status, KIP Data and Number of Siblings. Then classification identification was carried out from the data using the Random Forest, SVM, Naïve Bayes, Decision Tree, Neural Network algorithms in the Orange application. Based on the results of 433 research data that have been tested, precision, recall and accuracy calculation results are obtained for each model. the highest accuracy of Naïve Bayes and Random Forest is 95% (0,956). The results of this research show that the performance of Naïve Bayes and Random Forest is superior to SVM, Decision Tree and Neural Network. Decision Tree: (0.887), and SVM: (0.949) and Neural Network: (0.942).
