<![CDATA[Drop out is one of the problems faced by some students and can affect the quality of education in a university. Drop out occur due to various factors. One way to find out the factors that influence students to drop out is by classification. The classification of drop out in this study uses several variables, namely the Cumulative Achievement Index (GPA), Total Semester Credit Units, School Origin, Parents' Occupation, Parents' Income, Tuition Fees, Residence and Status. The classification algorithm used is the Decision Tree C4.5 Algorithm and the Naive Bayes Algorithm. This study was conducted to compare the performance of the Decision Tree C4.5 and Naive Bayes algorithms in order to obtain the best algorithm in classifying potential drop outs. The results obtained in this study are the Decision Tree C4.5 algorithm produces accuracy and f1-score values of 96.74% and 86.55%, while the Naive Bayes algorithm produces accuracy and f1-score values of 96.24% and 82.34. %. From these results, it can be concluded that the algorithm that has the best performance in classifying drop outs is the Decision Tree C4.5 Algorithm. The results of this research are implemented using a decision tree with a Total Semester Credit Unit as the root, which means the Total Credit is the most influential factor in the classification of drop outs. Recommendations for further research can be to compare more than two algorithms or add other variables in classifying students who have the potential to drop out.]]>
Copyrights © 2021
