@article{jitika, author = {Moh. Zainuddin}, title = { Perbandingan 4 Algoritma Berbasis Particle Swarm Optimization (PSO) Untuk Prediksi Kelulusan Tepat Waktu Mahasiswa}, journal = {Jurnal Ilmiah Teknologi Informasi Asia}, volume = {13}, number = {1}, year = {2018}, keywords = {}, abstract = {The purpose of this study was to find the best algorithm in making predictions of students' graduation from 4 algorithms: Naive Bayes Algorithm, Decision Tree (C4.5), k-Nearest Neighbor (kNN), Neural Network based Particle Swarm Optimization (PSO) as references to make policies and academic acts (BAAK) in reducing students who graduated late and did not pass. The results show that PSO-k-Nearest Neighbor (k-NN) algorithm based on k-optimum = 19 has the best performance of 4 algorithms, with Accuracy = 74,08% and Area Under the Curve (AUC) = 0,788. The addition of the Particle Swarm Optimization (PSO) feature always increases the accuracy value, where the highest accuracy value lies in the Decision Tree Algorithm (C4.5) of 5.21%, the lowest on the Naive Bayes Algorithm of 2.13%.}, issn = {2580-8397}, pages = {1--12}, url = {https://jurnal.stmikasia.ac.id/index.php/jitika/article/view/247} }