A Review of Students’ Graduation Classification: A Comparison of Naive Bayes Classifier and K-Nearest Neighbour

Abstract

Students are the most crucial aspects in determining the successful implementation of every program offered within educational institutions. Monitor the progress and students' achievement, enhance the ability of students, consider the number of students who have graduated, the ratio of the total number of students, the competence of graduates, are several major factors that require attention and serious consideration from the higher educational institutions. This study is mainly based on the data mining technique by implementing two common algorithms namely Naive Bayes Classifier and K-Nearest Neighbour due to classifying students' graduation (on time and overtime). The main objectives of this paper are to compare the achievement of both algorithms (NBC and KNN) towards students' graduation classification. This paper also focuses to identify the most important variables to predict students' performance. Leading to two categories of dependent variables namely graduate on time or graduate overtime. The considerations of these variables are based on the importance of each towards classifying the target. A Crossvalidation technique is applied to evaluate both algorithms. This study is beneficial for the center of graduate studies, educators, policymakers, and others in order to identify the main factors that impact the students' graduation status in higher educational institutions.

The Most Efficient Classifiers for the Students’ Academic Dataset

“…A comparison of classification algorithms has been applied in several fields like emotion classification, precipitation, Spatial modelling of storm dust provenance etc. Fauziastuti et al [22], classified students' graduation on time or overtime, by used two classification algorithms to compare their performance: Naive Bayes Classifier and K-Nearest Neighbor, but using a small dataset.…”