A Composer System based on Meta-Learning for Student Performance Prediction

The paper endeavors to augment the Manasa U Hegde efficiency and reliability of, the student performance prediction system and other analogous intelligence systems in Educational Data mining, by the amalgamation of multiple algorithms used for the purpose. Meta Learning and Ensemble methods are utilized to ameliorate the combined outcome in a multi class system. Additionally, a model is proposed, which is compliant even over external and noisy data, compared to the other individual algorithms used to conceive the composer system.