An Enhanced Evolutionary Student Performance Prediction Model Using Whale Optimization Algorithm Boosted with Sine-Cosine Mechanism

The students’ performance prediction (SPP) problem is a challenging that managers face at any institution. Collecting educational quantitative and qualitative data from many resources such as exam centers, virtual courses, e-learning systems, other not simple task. Even after collecting data, we might imbalanced missing biased different types strings, numbers, letters. One of the most common challenges in this area large number attributes (features). Determining highly valuable features needed to improve overall performance. This paper proposes an evolutionary-based SPP model utilizing enhanced form Whale Optimization Algorithm (EWOA) wrapper feature selection keep informative enhance quality. proposed EWOA combines (WOA) with Sine Cosine (SCA) Logistic Chaotic Map (LCM) WOA. SCA will empower exploitation process inside WOA minimize probability being stuck local optima. main idea worst half population using SCA. Besides, LCM strategy employed control diversity exploration process. As such, handled Adaptive Synthetic (ADASYN) sampling technique converting binary variant employing transfer functions (TFs) belong families (S-shaped V-shaped). Two real datasets are used, five classifiers employed: Decision Trees (DT), k-Nearest Neighbors (k-NN), Naive Bayes (NB), Linear Discriminant Analysis (LDA), LogitBoost (LB). obtained results show LDA classifier reliable both datasets. In addition, outperforms methods literature selected functions.