https://doi.org/10.31449/inf.v47i1.4519 Informatica 47 (2023) 11–20 11 Predicting Students Performance Using Supervised Machine Learning Based on Imbalanced Dataset and Wrapper Feature Selection Sadri Alija 1 , Edmond Beqiri 2* , Alaa Sahl Gaafar 3 , Alaa Khalaf Hamoud 4 1 Faculty of Business and Economics, South East European University, North Macedonia. 2 University of Peja “Haxhi Zeka” – Peja, Kosovo. 3 Department of Educational Planning, Directorate of Education in Basrah, Iraq. 4 Department of Computer Information Systems, University of Basrah, Iraq. Email: s.aliji@seeu.edu.mk, edmond.beqiri@unhz.eu, alaasy.2040@gmail.com, alaa.hamoud@uobasrah.edu.iq. Keywords: supervised machine learning, feature selection, wrapper, particle swarm optimization, info gain, SMOTE Received: November 14, 2022 For learning environments like schools and colleges, predicting the performance of students is one of the most crucial topics since it aids in the creation of practical systems that, among other things, promote academic performance and prevent dropout. The decision-makers and stakeholders in educational institutions always seek tools that help in predicting the number of failed courses for the students. These tools can help in finding and investigating the factors that led to this failure. In this paper, many supervised machine learning algorithms will investigate finding and exploring the optimal algorithm for predicting the number of failed courses of students. An imbalanced dataset will be handled with Synthetic Minority Oversampling TEchinque (SMOTE) to get an equal representation of the final class. Two feature selection approaches will be implemented to find the best approach that produces a highly accurate prediction. Wrapper with Particle Swarm Optimization (SPO) will be applied to find the optimal subset of features, and Info Gain with ranker to get the most correlated individual features to the final class. Many supervised algorithms will be implemented such as (Naïve Bayes, Random Forest, Random Tree, C4.5, LMT, Logistic, and Sequential Minimal Optimization algorithm (SMO)). The findings show that the wrapper filter with SPO-based SMOTE outperforms the Info-Gain filter with SMOTE and improves the performance of the algorithms. Random Forest outperforms the other supervised machine learning algorithms with (85.6%) in TP average rate and Recall, and (96.7%) in ROC curve. Povzetek: 1 Introduction High-quality universities always require a great record of their students and the students are the main resource for them. The main concern for the universities is the performance of the students which is the base stone for building the top rate graduates and post-graduate students who will be the leaders of the nations and take responsibility of the economic and social growth of the society. Moreover, the main concerns for market employers are the performance of universities and students’ academic performance due to its direct effect on the employment process and then employee productivity. So, the employers’ demands are met by the graduated students who exert efforts in their academic journey. Student performance is measured by the learning assessment and the curriculum according to Usamah et al [1]. It is frequently important to be able to predict the behavior of future students to enhance the design of the curriculum and prepare the interventions for academic guidance and support. Machine learning (ML) is useful in this situation. ML approaches examine datasets, extract information, and then organize that information for eventual use. The primary goals of ML are to identify and extract patterns from recorded data by using a variety of techniques and algorithms [2]. Numerous algorithms exist and are used with educational data, including supervised algorithms such as Decision Tree (DT) and Naive Bayes (NB), and unsupervised algorithms such as K-Nearest Neighbor (KNN), and Neural Network (NN). Such algorithms forecast patterns, upcoming trends, and behaviors, enabling businesses to make informed, proactive decisions mining. This paper's major goal is to predict student performance using Supervised ML based on an imbalanced dataset and wrapper feature selection. The following section sheds light on related previous studies, then followed by the methodology and the concluded points, and future work.