CD Skripsi
Implementasi Metode Random Forest Dalam Menentukan Kelulusan Beasiswa Kartu Indonesia Pintar Kuliah Di Universitas Riau
The Indonesia Smart College Card (KIP-K) scholarship is one of the scholarships that is in high demand by Riau University students. Analysis to determine objective scholarship recipients is needed. In addition, it is also necessary to identify factors that affect KIP-K graduation. This research builds a prediction model for KIP-K scholarship graduation at Riau University using the Random Forest algorithm. The data used are 2,673 students of class 2022 of Riau University who registered themselves to get KIP-K. In this research, cross validation technique is used to measure the performance of the model. The model accuracy results were evaluated using confusion matrix, and it was found that the Random Forest model with 25 ntree gave an average accuracy of 76.3%, precision of 76.54% and recall of 73.56%. Factors that are closely related to KIP-K graduation are coming from a family registered in the Targeted Program for Accelerating the Elimination of Extreme Poverty (P3KE) and KIP ownership status at the previous education level. With the results of accuracy, precision and recall that are quite high, the Random forest method can be declared feasible to use in predicting KIP-K scholarship graduation at Riau University.
Keywords : Cross Validation, Confusion Matrix, Dictionary, KIP-K, Random
Forest.
Tidak tersedia versi lain