Optimasi Seleksi Fitur Information Gain pada Algoritma Naïve Bayes dan K-Nearest Neighbor

There was an increase in the number of late payments tuition fees by 3,018 from a total 5,535 students at end 2020. This study uses Python library which requires data to be numeric type, so it transformation according type study, that has scale is transformed using ordinal encoder, and does not have one-hot encoding. The purpose this evaluate performance Naïve Bayes algorithm K-Nearest Neighbor with confusion matrix predicting payment UMKT. dataset used sourced financial administration bureau as many 12,408 distribution 90:10. Based on results calculation selection information gain features, best 4 attributes influence research are obtained, namely faculty, program, class, gender. evaluation obtain accuracy 55.19%, while only obtains 50.76%. obtained prediction derived gain, influences increasing Bayes, but use attribute makes decrease.