On Analysis and Evaluation for Predicting Students’ Academic Performance GPA Considering an Engineering Institution (Neural Networks’ Modeling Approach)

Abstract

Predicting students’ performance is one of the most important topics for learning contexts such as schools and universities, since it helps to design effective mechanisms that improve academic results. Educational Institutions face numerous challenges today in providing quality and student-centric education to Students Individual learners prefer their own strategies originated from diverse learning styles. Learning style models may include collective strategies for mental, emotional, and physiological components. On the basis of such components, this piece of research suggests a specific quantified learning style preferred by learners in engineering education. By following average learners’ achievements (marks) at specific courses closely related to the specialization, interesting analytical results for Grade Point Average (GPA) evaluation are obtained. Moreover, an ANN model with supervised learning is presented to simulate diverse learning styles performance. Accordingly, optimal guided advise is suggested in fulfillment of probabilistically best GPA of graduated engineers. Obtained simulation results are well supported by the findings of experimental case study.