Determining the impact of demographic features in predicting student success in Croatia

Trstenjak, Bruno; Ðonko, Dženana

doi:10.1109/mipro.2014.6859754

Cited by 19 publications

(10 citation statements)

References 8 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…As examples of (1), some researchers [40,47] assess the quality of their regression models using the models themselves. Exemplifying (2), other authors [11,51] generate predictive models and provide prediction results. However, those authors test their models on a test set pulled from the same cohort on which training occurs, which may cause the models to overfit to that cohort.…”

Section: Institution-and Program-level Modelingmentioning

confidence: 99%

See 1 more Smart Citation

A Robust Machine Learning Technique to Predict Low-performing Students

Liao

Zingaro

Thai

et al. 2019

ACM Trans. Comput. Educ.

View full text Add to dashboard Cite

As enrollments and class sizes in postsecondary institutions have increased, instructors have sought automated and lightweight means to identify students who are at risk of performing poorly in a course. This identification must be performed early enough in the term to allow instructors to assist those students before they fall irreparably behind. This study describes a modeling methodology that predicts student final exam scores in the third week of the term by using the clicker data that is automatically collected for instructors when they employ the Peer Instruction pedagogy. The modeling technique uses a support vector machine binary classifier, trained on one term of a course, to predict outcomes in the subsequent term. We applied this modeling technique to five different courses across the computer science curriculum, taught by three different instructors at two different institutions. Our modeling approach includes a set of strengths not seen wholesale in prior work, while maintaining competitive levels of accuracy with that work. These strengths include using a lightweight source of student data, affording early detection of struggling students, and predicting outcomes across terms in a natural setting (different final exams, minor changes to course content), across multiple courses in a curriculum, and across multiple institutions.

show abstract

Section: Institution-and Program-level Modelingmentioning

confidence: 99%

“…[17] Institution [40] Institution [47] Institution [11] Institution [51] Institution [21] Program [44] Program [1] Program [19] Program [42] Program [27] Course [41] Course [53] Course [6] Course [20] Course [52] Course [8] Course [4] Course [24] Course [2] Course [10] Course…”

mentioning

confidence: 99%

A Robust Machine Learning Technique to Predict Low-performing Students

Liao

Zingaro

Thai

et al. 2019

ACM Trans. Comput. Educ.

View full text Add to dashboard Cite

show abstract

“…In general, there are number of studies in this research process have been accomplished that uses various methodologies and techniques for the evaluation of student's performance. These comprise Artificial Neural Networks (ANN) [21], decision tree algorithms [27], Naive Bayes [28] have used "key" democratic variables of students and their academic grade for the performance prediction of students in open university based on six different algorithms of linear regression, neural networks, model trees and support vector machine. A hybrid algorithm has been implemented in [3] by using the concept of clustering and decision tree algorithm for the classification of the data samples.…”

Section: Related Workmentioning

confidence: 99%

A model for improved performance prediction using ensemble-based hybrid classification approach on a multivariate student dataset

Anoopkumar

Rahman

2018

IJET

View full text Add to dashboard Cite

Classification techniques have sensed substantial attention in Information Engineering and Technology for the performance prediction and optimisation since few decades. The discovered accuracy of the Classification Model helps the institutional practices and student's performances. In this paper, a novel Ensemble-based Hybrid Classification Approach (EHCA) has been proposed to be managed to produce improved performance prediction. The mining process with new attributes based on student behaviours has also been incorporated, since it creates a great impact on their academic performances. Moreover, the performance of the students is analysed with a set of classifiers in Educational Data Mining (EDM) namely, Naive Bayesian, Support-Vector-Machine (SVM) and J48. Futuristic-bound Ensemble approach is employed for enhancing the classifier performances. Here, the futuristic methods of ensembles of Bagging, Classification Boosting and Stacking are used for optimising the results with more precision. Further, the process of Ensemble-based Hybrid Classification is analysed and tested with the dataset collected from Kerala Technological University-SNG College of Engineering (KTU_SNG). The results obtained are compared with the results obtained for utilized single classifiers and the EHCA on the basis of performance efficiency and classification accuracy. The work evidence the efficiency of the pro-posed approach and proves its reliability in Profound Performance Prediction and Optimisation.

show abstract

“…A common approach for building such prediction models is to train a machine-learning algorithm with student performance data collected from the final years of high school combined with students' demographic data. It has been shown such data are indeed good predictors of success at a critical point such as the end of the first year of college studies ( [6], [27], [28], [29], [30], [31], [32], [33]). However, much of the demographic data seldom changes and academic performance history from earlier years never does.…”

Section: One-off Versus Continuous Prediction -The Case Of Summative mentioning

confidence: 99%