A COMPARATIVE STUDY OF MACHINE LEARNING ALGORITHMS FOR PREDICTING STUDENT PERFORMANCE

2.1.6 K-Nearest Neighbors (KNN): A simple algorithm that predicts the target variable based on the values of its neighbors. KNN is often used for both classification and regression tasks (Otu et al., 2024) (Qureshi & Lokhande, 2024). KNN is a simple algorithm that predicts the target variable based on the values of its neighbors. It has been used in various educational data mining tasks, including predicting student performance. KNN has been shown to achieve high accuracy rates, with one study reporting an accuracy of 88% (Otu et al., 2024). However, KNN can be computationally expensive for large datasets and may not handle high-dimensional data effectively 2.1.7 Gradient Boosting and XGBoost: These are advanced ensemble learning algorithms that combine multiple weak models to create a strong predictive model. They are known for their high accuracy and are widely used in educational data mining (Otu et al., 2024) (Airlangga, 2024). Gradient Boosting and XGBoost are advanced ensemble learning algorithms that combine multiple weak models to create a strong predictive model. They have been shown to achieve high accuracy rates in predicting student performance, with one study reporting an accuracy of 90% (Otu et al., 2024). However, these algorithms can be computationally expensive and may require careful tuning of hyperparameters. 2.2 Educational Data Mining and Student Performance Prediction Educational Data Mining (EDM) is an emerging field that applies data mining techniques to educational data to uncover patterns and insights. One of the primary applications of EDM is the prediction of student performance, which helps educators identify at-risk students and optimize learning strategies