Enriched model of Students Academic Performance Prediction based on their Learning Life Style using improved Fuzzy Min-Max Neural Classifier with Firefly Algorithm

Abstract

The foremost challenge in Education institutions is providing quality education is a sustainable development in the field of education. Predicting the academic performance of the college students is an essential task to enhance the capability of students by providing them various mode of teaching methodologies. But predicting their performance very accurately is a toughest task while using a standard statistical model, thus education data mining plays a vital role in discovering useful pattern of student information. This paper, mainly concentrates on the learning behavior of the students as the major factor along with considering other attributes of student’s academic details. Handling of voluminous student dataset is greatly handled by applying tuned relief feature subset selection which reveals the most prominent attributes which contributes more in the prediction process and increase the relevancy of the entire process. Optimized Fuzzy Min Max Neural Classifier is used to handle the inconsistency of the prediction process and it achieves its efficacy by adapting firefly algorithm to fine tune the parameters of the network, so that its performance result proves its benefit than the other existing prediction models.