Predicting students’ academic achievement: Comparison between logistic regression, artificial neural network, and Neuro-fuzzy

Modeling Course Achievements of Elementary Education Teacher Candidates with Artificial Neural Networks

“…This leads to different analyses and different outcomes in similar studies, as it is the case in the prediction of academic achievement. However, it is seen that the lowest classification prediction and the highest network performance vary between 51.88% and 91.77%, respectively, in the findings of studies (such as linear regression, logistic regression and decision tree) on the comparison of the predictions and the statistical methods with respect to artificial neural networks related to education and training (Bahadır, 2013;Çırak, 2012;Gülçin, Çırak, & Çokluk, 2013;Demir, 2015;Guo, 2010;Ibrahim & Rusli, 2007;Karamouzis & Vrettos, 2008;Kardan, Sadeghi, Ghidary, & Sani, 2013;Moridis & Economides, 2009;Naser, Zaqout, Ghosh, Atallah, & Alajrami, 2015;Oancea, Dragoescu, & Ciucu, 2013;Oladokun, Adebanjo, & Charles-Owaba, 2008;Paliwal & Kumar, 2009;Romero, Ventura, & García, 2008;Rusli, Ibrahim, & Janor, 2008;Şen, Uçar, & Delen, 2012;Şengür, 2013;Şengür & Tekin, 2013;Tepehan, 2011;Tosun, 2007;Turhan, Kurt, & Engin, 2013;Vandamme, Meskens, & Superby, 2007).…”

Predicting Achievement with Artificial Neural Networks: The Case of Anadolu University Open Education System

“…When the studies on the prediction of student achievement are examined, it is seen that in general, different mathematical models such as Regression, Artificial Neural Networks, Decision support Systems, Decision Trees and Baynes were compared and that ANN displayed a better performance than others (Herzog, 2006;Lykourentzou et al, 2009;Naik and Ragothaman, 2004;Schumacher, Olinsky, Quinn, & Smith, 2010;Şen, Uçar & Delen, 2012;Rusli, Ibrahim & Janor, 2008;Turhan et al, 2013). As distinct from these results, a practice with ANN was ranked the second in the study carried out by Aydın (2007) with an accuracy rate of 77.80%; and ANN analysis was ranked the third in the study carried out by Yükseltürk et al with a classification rate of 79.7%.…”

“…Turhan et al, 2013 ANN (best result) and regression analyses Lykourentzou et al, 2009 ANN displayed more effective performance, in comparison to linear regression. Aydın, 2007 C5.0, Logistic Regression, ANN, C&RT, CHAID and QUEST Rusli et al, 2008 Prediction through ANN is provided more exact results than decision tree and linear regression. Naik & Ragothaman, 2004 Prediction through ANN is 93.38% exact.…”

An Integrated System Framework for Predicting Students' Academic Performance in Higher Educational Institutions

“…Moreover, it should be noted that these factors that affect and determine student performance are not solitary in nature but are interconnected, interrelated and interdependence (Figure 2). [25 ] [26 ] and [27 ] suggested that there is a possibility to improve the student prediction accuracy with the used of more independent variables or attributes that are outside the database of the University system. Therefore, there is a need for research to develop a new framework that is comprehensive and holistic in its approach.…”