KNN and Naive Bayes for Optional Advanced Courses Recommendation

This study focuses on the use of KNN and Naive Bayes to predict students on taking optional advanced courses that most of the time lead them to fail get good score and postpone their graduation. The purposes of this study are to find out how KNN and Naive Bayes algorithms can be used to recommend the best option of advanced courses for students and to know the accuracy of the algorithm. The data used for this study is collected from academic data from Computer Education program students in Mulawarman University. Independent variables are scores of the basic courses of the previous class year students and dependent variable is recommendation of advanced courses based on the history of advanced courses scores with study period less than 5 years. The result of this research shows that accuracy score of KNN algorithm is 83,33% with value of K=2, accuracy score of KNN algorithm is 100% with value of K=1, and accuracy score of Naive Bayes algorithm is 100%. Although the data is still relatively small, it is quite good for preparatory to build a recommender system of advanced courses for students.