Prediction of Subject Requests in Short Semester by Using Engineering Association Rules with Apriori Algorithms at Fakultas Teknologi Informasi Universitas Advent Indonesia

Abstract

Short semester is a period of lectures carried out by an educational institution / university with the aim of facilitating students who wish to shorten their lecture time, as well as those who wish to improve their grades that are inadequate.  In this semester,  the preparations made refer to the student's request for certain  courses  that he wants to attend.  In order to make a good  preparation, the faculty leaders need to predict what courses are likely to be taken by students, so that they can prepare the registration and assign the lecturers  whose related to the course. The purpose of this study is to collect data of students request for certain courses useing previous data.  For data mining technique, the author  used the Association Rule and apriori algorithm to find the frequent itemset  to make prediction. By analyzing 25 participant data of students at  the Faculty of Information Technology, on short semester of the academic year 20162017, the resuts shows that there was a high level of support and confidence in several combinations of subjects. In addition, it is also found that the rule with the highest final association value is in the rule of "Choose B then choose K", namely: If the student chooses course B (Algorithm), then he will also choose course  K (Algorithm  Practice) with the value of support 20% and 100% confidence.

Keywords: Data Mining, Association Rule, Apriori Algorithm