TY - JOUR AU - Noripansyah, Noripansyah AU - Kadir, Abdul AU - Kusumaningsih, Dewi AU - Haderiansyah, Haderiansyah PY - 2023/06/26 Y2 - 2025/11/15 TI - ACCREDITATION PREDICTION OF EARLY CHILDHOOD EDUCATION INSTITUTIONS USING MACHINE LEARNING TECHNIQUES JF - Jurnal Teknik Informatika (Jutif) JA - J. Tek. Inform. (JUTIF) VL - 4 IS - 3 SE - Articles DO - 10.52436/1.jutif.2023.4.3.999 UR - https://jutif.if.unsoed.ac.id/index.php/jurnal/article/view/999 SP - 647-655 AB - <p>Accreditation is an acknowledgement of an educational institution regarding the feasibility of carrying out the educational process. Making predictions can save time for early childhood education institutions in compiling accreditation forms that will be submitted. Prediction in determining accreditation becomes an important lesson for an institution in self-assessing the quality of its services. Choosing which method to use in the accreditation prediction process becomes a serious problem, so the prediction results can be the closest or most accurate. Machine Learning is an application that is part of Artificial Intelligence which is widely used in prediction research. In this experiment, three algorithms in machine learning are tested, namely SVM, KNN and ANN. This study uses data from the accreditation results of early childhood education institutions in South Kalimantan; the sample data is 75%, and the remaining data is 25%. The results of the KNN algorithm with Euclidean distance and the number of neighbours 5 have the best performance in predicting the value of the accreditation predicate compared to other methods. The results of calculations using the KNN method produce Area Under Curve values of 1,000, CA 1,000, F1 1,000, precision 1,000 and Recall 1,000.</p> ER -