Machine learning prediction and analysis of students’ academic performance - Research on engineering student study success predictors

Aug 3, 2026

Upload by Levi

Student academic performance analysis supports enrollment decision-making and clarifies standards for secondary-school graduates entering higher education. This research builds a machine learning Decision Tree classification model to predict engineering students’ outcomes using secondary education records. Performance is measured as a binary result: successful completion of the first and second study years. Predictor variables include general success, competition awards, subject grades in mathematics and physics, one official state language, plus transferred-course counts and first-year GPA. Results show different key drivers across first- and second-year enrollment.

LanguageEnglish

CategoryResearch & Report

FormatPDF

Pages20

Reading Time50 min

AccessFree to download

Introduction

Literature review on SAP prediction methods

Categories of predictive models and performance indicators

Factors used for academic performance prediction

Notable related studies

Machine learning prediction and analysis of students’ academic performance Mirza Pasic1, Ajdin Vatres1, Faris Ferizbegovic1, Hadis Bajric1, Mugdim Pasic1 1 Department of Industrial Engineering and Management, Faculty of Mechanical Engineering, University of Sarajevo, Bosnia and Herzegovina | ABSTRACT | | | --- | --- | | Analyzing students’ academic performance is important for evaluating enrollment criteria which establish the standards required for pupils who finished secondary school to gain admission to a higher education institution. The aims of this research were to develop a machine learning prediction Decision Tree classification model and analyze the performance of engineering students based on their performances during secondary school education. The performance of students was analyzed and measured as a binomial response whether students successfully finished the first and the second study years. The developed model examined general success, number of awards obtained at competitions, special awards, average grades in mathematics, physics, and one of the official state languages during secondary school as predictor variables. The number of courses transferred from the first into the second study year and students’GPA obtained during the first study year were added as predictor variables in the analysis and development of a prediction model for the students’ performance during the second study year and their enrollment in the third study year. Developed machine learning prediction model showed that for the performance of enrolled students in the first study year general success of students during secondary school is the most important predictor variable, followed by mathematics and physics grades. However, for the performance of the students enrolled in the second study year the most important predictor variable was number of the courses transferred from the first into the second study year, followed by students’GPA obtained during the first study year and general success. Machine learning Decision Tree classification modeling was shown to be an adequate tool for the prediction of the performance of engineering students during the first and second study years. | | | Keywords: | Machine learning, Decision Tree, Enrollment criteria, Engineering students, Study success | | Corresponding Author:
Mirza Pasic
Department of Industrial Engineering and Management Faculty of Mechanical Engineering
University of Sarajevo Vilsonovo setaliste 9
[E-mail: mirza.pasic@mef.unsa.ba](E-mail: mirza.pasic@mef.unsa.ba) | | 1. Introduction The frequency at which various statistical and machine learning methods have been used to predict student performance was analyzed in a systematic literature review of 357 articles on predicting students’ academic performance (SAP) using various statistical and machine learning methods. It was concluded that 31.3% of authors are using statistical methods such as linear regression and ANOVA with Decision Trees [1] . Clustering methods [1] are less used than classification techniques and clustering is mostly used as a preparation for applying the model. Following this in [2] a further literature survey is performed, shedding more light on what types of models are most frequently used for SAP prediction. The most common models in use are based on Decision Tree, Naïve Bayes, and Rule-Based algorithms, with GPA, gender, age, and marital status as factors. In [3] methods for predicting student performance are divided into four categories Regression, Clustering, Decision Tree, and Dimensionality reduction. These methods are used to predict many different student performance indicators including course grade or score, grade point average (GPA) or range of GPA, additionally cumulative grade point average (CGPA), semester grade point average (SGPA), course retention or dropout, program or module graduation, and more. For example, in [4]-[6] course grades or scores are predicted, in [7]

FAQs

What student outcomes does the model aim to predict?

It predicts whether engineering students successfully finish the first and second study years, using a binary response definition of study-year completion.

Which variables are used as predictors in the research?

Predictors include general success, numbers of awards at competitions and special awards, average grades in mathematics, physics, and an official state language, plus the number of courses transferred into the next year and students’ first-year GPA.

How do the most important predictors differ between the first and second study years?

For first-year performance, general success during secondary school is the strongest predictor, followed by mathematics and physics grades. For second-year performance, the number of courses transferred from year one to year two is most important, followed by first-year GPA and general success.

Aug 3, 2026

Upload by Levi