Enhancing Academic Success: A Novel Approach to Predict Learning Performance with an Advanced Blended Learning Performance Predictor

Abstract

The concept of educational data mining focuses on the development of methods for investigating and analyzing the enormous amounts of information that are produced by the academic environment. It is in a position to capitalize on the large number of studies produced by the data mining industry and connect that knowledge to academic issues pertaining to teaching, reasoning, and evaluation. Recently, this sector has become predominantly effective at solving various difficulties with scholastic analytics. This achievement can be attributed to the vast processing capacity and information retrieval methods that are used. The primary goal of educational establishments at the university level is to furnish the students who attend those institutes with a curriculum of sufficient calibre. The pursuit of information that can accurately anticipate students’ performance is one strategy for elevating the standard of the educational structure at the postsecondary level to the greatest possible standard. In this study, a superior Blended Learning Performance Predictor Toolkit (BLPPT) is presented that can be used to predict how students would perform on their semester examinations utilizing the data collected through surveys. The BLPPT model reaches an MAE score of 2.94 × 10¹¹ and an MSE value of 9.18 × 10²³.