Prediction of Student Satisfaction with Lecturer Performance in the Informatics Engineering Study Program at Ngudi Waluyo University Using the C4.5 Algorithm

Abstract

The rapid development of the times has made the world of education face very tough challenges. Very tight competition makes every college must pay more attention and improve the quality of education including the quality of lecturers as the party that produces or provides educational services to students. Lecturers who have good performance can also increase students' enthusiasm for learning and improve the quality of their learning outcomes. For this reason, predicting student satisfaction with lecturer performance is very important for development. This research was conducted with data mining techniques using the C4.5 algorithm with the aim of knowing the accuracy level of the C4.5 algorithm and in order to help universities to find out student satisfaction with lecturers so that they can continue to improve the quality of academic services received by their students. The C4.5 algorithm is one of the data mining classification methods whose results are in the form of decision tree patterns/models and decision rules. This research produces accuracy results of 95.26%, class recall of 98.09%, and class precision of 96.86%. These results show high accuracy because the rules or rules produced are close to 100% and the C4.5 algorithm can predict data correctly and close to the actual value.