| Summary: | In today's highly competitive educational landscape, it has become essential for institutions to proactively forecast student outcomes, categorize individuals based on their abilities, and actively work towards enhancing their performance in upcoming assessments. Providing students with timely guidance is crucial, steering them towards concentrating their efforts on specific areas to elevate their academic achievements. The analysis of student and teacher data plays a pivotal role in achieving these goals. This analytical approach not only aids institutions in reducing failure rates but also enables them to predict students' performance in a course by leveraging insights from their past achievements in similar classes. At the heart of this process is data mining, a collection of methods employed to unearth concealed patterns within extensive datasets. These patterns, once revealed, hold significant potential for analysis and prediction. These applications focus on a comprehensive examination of data sourced from both students and teachers. The primary objectives of this analysis are manifold, encompassing tasks such as classification and prediction. To this aim, the model used in this study is Random Forest Classification (RFC). To enhance the model, this study employed two optimizers, namely, the Artificial Hummingbird Algorithm (AHA) and the Flow Direction Algorithm (FDA). This model can be used in the real world for prediction of student performance. The best model in performance in this study is the RFAH model. The accuracy value of the RFAH model is (0.954), which is the best performance by far in this study.
|
|---|
