Predicting student's dropout in university classes using two-layer ensemble machine learning approach: A novel stacked generalization

Niyogisubizo, Jovial; Liao, Lyuchao; Nziyumva, Eric; Murwanashyaka, Evariste; Nshimyumukiza, Pierre Claver

doi:10.1016/j.caeai.2022.100066

Cited by 64 publications

(39 citation statements)

References 39 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…Furthermore, the adoption of Information and Communication Technologies (ICT) in the educational system can be perceived as a window of opportunity for the optimization of didactic processes, valuing the options available to improve student adherence and engagement in the pedagogical process (Niyogisubizo et al, 2022). Similarly, technological tools facilitate the improvement of student engagement through positive feedback on student performance, as well as contributing to the early identification of unmet educational needs that can be addressed by the education system.…”

Section: Discussionmentioning

confidence: 99%

Multicausal analysis of the dropout of university students from teacher training studies in Andalusia

Ibáñez-Cubillas

López-Rodríguez²,

Sánchez³

et al. 2023

Front. Educ.

View full text Add to dashboard Cite

To give an answer and a solution to the inconveniences that dropout brings to universities and to society, we have studied the different approaches that exists in the actual scientific literature about the reasons of this phenomenon. The aim of this investigation is analyzing the college dropout, focalizing in students of education degrees of Andalusian universities. For doing that, we applied a standardized instrument that pretend to determine which are the factors associated to the permanence of the college students. The instrument was applied to 608 students of the first year of six Andalusian universities. Of these, 274 were studying Infant education degree and 334 Primary education degree. The study shows that a large sample of the subjects interviewed responded with higher scores 3.83, stating the need to remain in their undergraduate studies, with few references to dropping out of the university. Some of the items that have obtained the highest ratings express their desire to graduate with honors, become a good professional and practice their profession within their field of work, among others.

show abstract

Section: Discussionmentioning

confidence: 99%

Multicausal analysis of the dropout of university students from teacher training studies in Andalusia

Ibáñez-Cubillas

López-Rodríguez²,

Sánchez³

et al. 2023

Front. Educ.

View full text Add to dashboard Cite

show abstract

“…Refs. [21][22][23][24][25][26][27][28] predicted the students' performance and whether or not the student will drop out. Researchers in Ref.…”

Section: Predicting Student Dropout Authors Inmentioning

confidence: 99%

“…Moreover, a group of researchers in Ref. [22] proposed a stacking ensemble model that combines RF, eXtreme gradient boosting (XGBoost), gradient boosting (GB), and artificial neural networks (ANN) to predict the student that may be at risk of being a drop out at an individual course. In this study, we used 261 students' samples and 12 attributes collected from 2016 to 2020 at Constantine the Philosopher University in Nitra.…”

Section: Predicting Student Dropout Authors Inmentioning

confidence: 99%

Predicting Student Academic Performance at Higher Education Using Data Mining: A Systematic Review

Alwarthan

Aslam

Khan

2022

Applied Computational Intelligence and Soft Computing

View full text Add to dashboard Cite

Recently, educational institutions faced many challenges. One of these challenges is the huge amount of educational data that can be used to discover new insights that have a significant contribution to students, teachers, and administrators. Nowadays, researchers from numerous domains are very interested in increasing the quality of learning in educational institutions in order to improve student success and learning outcomes. Several studies have been made to predict student achievement at various levels. Most of the previous studies were focused on predicting student performance at graduation time or at the level of a specific course. The main objective of this paper is to highlight the recently published studies for predicting student academic performance in higher education. Moreover, this study aims to identify the most commonly used techniques for predicting the student's academic level. In addition, this study summarized the highest influential features used for predicting the student academic performance where identifying the most influential factors on student’s performance level will help the student as well as the policymakers and will give detailed insights into the problem. Finally, the results showed that the RF and ensemble model were the most accurate models as they outperformed other models in many previous studies. In addition, researchers in previous studies did not agree on whether the admission requirements have a strong relationship with students' achievement or not, indicating the need to address this issue. Moreover, it has been noticed that there are few studies which predict the student academic performance using students’ data in arts and humanities major.

show abstract

“…The designed model prediction accuracy was not sufficient. A new stacking ensemble based on a hybrid of random forest (RF), extreme gradient boosting (XGBoost), gradient boosting (GB), and feed‐forward neural networks (FNNs) was introduced in Reference 25. Logit leaf model (LLM) algorithm was developed in Reference 26 with higher student dropout predictions.…”

Section: Related Workmentioning

confidence: 99%

Rank biserial stochastic feature embed bivariate kernelized regressive bootstrap aggregative classifier for school student dropout prediction

Chinnasamy

Balasubramanian

2022

Concurrency and Computation

View full text Add to dashboard Cite

Early and accurately predicting the students' dropout enables schools to recognize the students based on available educational data. The early student dropout prediction is a major concern of education administrators. The existing classification techniques were unable to handle the early stage accurate performance of student dropout prediction with maximum accuracy and minimum time. In order to resolve the issue, a novel technique called rank biserial Otsuka-Ochiai stochastic embedded feature selection based bivariate kernelized regressive bootstrap aggregative classifier (RBOOSEFS-BKBAC) is motivated to perform student dropout prediction. The aim of the designing RBOOSEFS-BKBAC is to improve student dropout accuracy and minimal time consumption. Initially, the data preprocessing is to perform the data normalization, data cleaning, and duplicate data removal. Next, rank biserial correlation is used for discovering the correlated features. Followed by, Otsuka-Ochiai stochastic neighbor embedded feature selection is carried out to select significant features. Finally, bivariate kernelized regressive bootstrap aggregative classification technique is to perform classification with help of weak classifier. By using Bucklin voting scheme, the classification outcomes are obtained for increasing prediction accuracy as well as minimizing error. Experimental evaluation is performed by using Student-Drop-India2016 dataset with different metrics such as prediction accuracy, precision, recall, F-measure, as well as time. The result of proposed RBOOSEFS-BKBAC technique is provided that the higher prediction accuracy by 5% and lesser the prediction time by 15%, as compared to the state-of-the-art methods. K E Y W O R D Sbivariate kernelized regressive bootstrap aggregative classification, Otsuka-Ochiai stochastic neighbor embedded feature selection, preprocessing, rank biserial correlation-based feature extraction, school student dropout prediction INTRODUCTIONStudents' dropout is a difficult issue in the learning process for both the individual and society. Early prediction is a main challenge in the dropout prediction. Predicting the student's at-risk of dropout offers various benefits to educators, students, and the educational institution. Machine learning algorithms were applied in education to address the predicting dropout of students issue from schools. So that potential drop outs are documented

show abstract

Predicting student's dropout in university classes using two-layer ensemble machine learning approach: A novel stacked generalization

Cited by 64 publications

References 39 publications

Multicausal analysis of the dropout of university students from teacher training studies in Andalusia

Multicausal analysis of the dropout of university students from teacher training studies in Andalusia

Predicting Student Academic Performance at Higher Education Using Data Mining: A Systematic Review

Rank biserial stochastic feature embed bivariate kernelized regressive bootstrap aggregative classifier for school student dropout prediction

Contact Info

Product

Resources

About