<![CDATA[To assure the quality of graduates, it is required to estimate the graduation rate of active students based on variables that influence it, such as first-through-sixth-semester GPAs, the number of credits taken each semester, etc. Graduation rate is a criterion for evaluating the accreditation of study programs and institutions, making it one of the benchmarks for higher education management policies. In order to forecast student graduation rates, an artificial neural network algorithm based on the Adaptive Neuro-Fuzzy Inference System approach was used to analyze data in this study. This technique is commonly employed for problem prediction. In the implementation of this technique, the sample data consist of around 627 student data from the classes of 2015 through 2018. With the result that predicts the number of years and months till student graduation. Good accuracy results were obtained with the approach utilized, which included the kind of membership function, namely gauss mf, gbell mf, trim mf, and traf mf. On average, it provided a R value of 0.99 at epoch values between 50 and 500, an MSE value of 0.04, and an accuracy rate of 96.97%]]>
Copyrights © 2023
