Prediction of student performance using machine learning techniques

The ever-increasing importance of education has driven researchers and educators to seek innovative methods for enhancing student performance and understanding the factors that contribute to academic success. This paper presents a methodology for predicting student performance (SPP) that leverages machine learning techniques to forecast students' academic achievements based on a variety of features, such as demographic information, academic history, and behavioral patterns. The proposed SPP method utilizes a real-world collected dataset from multiple educational institutions to ensure an accurate and comprehensive analysis.The proposed methodology starts with a data preparation stage, where the data is cleansed and organized for analysis. This process encompasses tasks such as handling missing values, scaling the data, and transforming variables if necessary. The RFE selection technique was used to select the most important features for the SPP model. A number of machine learning classifiers were tested, and the linear regression algorithm was found to be the best performer. The evaluation results showed that the linear regression algorithm can be used to predict the student performance with a mean absolute error (MAE) of 0.23 and a root mean square error (RMSE) of 0.29.The results of this study demonstrate the potential of machine learning algorithms in predicting student performance and identifying key factors that influence academic success. This information can be leveraged by educators and academic institutions to develop targeted intervention strategies, tailored learning experiences, and personalized recommendations for students, ultimately fostering a more effective learning environment and improving overall educational outcomes.