Prediction of Academic Performance Applying NNs: A Focus on Statistical Feature-Shedding and Lifestyle
Journal Title: International Journal of Advanced Computer Science & Applications - Year 2019, Vol 10, Issue 9
Abstract
Automation has made it possible to garner and preserve students’ data and the modern advent in data science enthusiastically mines this data to predict performance, to the interest of both tutors and tutees. Academic excellence is a phenomenon resulting from a complex set of criteria originating in psychology, habits and according to this study, lifestyle and preferences–justifying machine learning to be ideal in classifying academic soundness. In this paper, computer science majors’ data have been gleaned consensually by surveying at Ahsanullah University, situated in Bangladesh. Visually aided exploratory analysis revealed interesting propensities as features, whose significance was further substantiated by statistically inferential Chi-squared (χ2 independence tests and independent samples t-tests for categorical and continuous variables respectively, on median/mode-imputed data. The initially relaxed p-value retained all exploratorily analyzed features, but gradual rigidification exposed the most powerful features by fitting neural networks of decreasing complexity i.e., having 24, 20 and finally 12 hidden neurons. Statistical inference uniquely helped shed off weak features prior to training, thus optimizing time and generally large computational power to train expensive predictive models. The k-fold cross-validated, hyper-parametrically tuned, robust models performed with average accuracies wavering between 90% to 96% and an average 89.21% F1-score on the optimal model, with the incremental improvement in models proven by statistical ANOVA.
Authors and Affiliations
Shithi Maitra, Sakib Eshrak, Md. Ahsanul Bari, Abdullah Al-Sakin, Rubana Hossain Munia, Nasrin Akter, Zabir Haque
Keywords
Related Articles
- EP ID EP646217
- DOI 10.14569/IJACSA.2019.0100974
- Views 92
- Downloads 0
Influence of Adopting a Text-Free User Interface on the Usability of a Web-based Government System with Illiterate and Semi-Literate People
Illiterate and semi-literate people usually face different types of difficulties when they use the Internet, such as reading and recognising text. This research aims to develop and examine the influence of adopting a tex...
Divide and Conquer Approach for Solving Security and Usability Conflict in User Authentication
Knowledge based authentication schemes are divided into textual password schemes and graphical password schemes. Textual password schemes are easy to use but have well known security issues, such as weak against online s...
Novel Software-Defined Network Approach of Flexible Network Adaptive for VPN MPLS Traffic Engineering
Multi-Protocol Label Switching VPN (MPLS-VPN) is a technology for connecting multiple remote sites across the operator’s private infrastructure. MPLS VPN offers advantages that traditional solutions cannot guarantee, in...
SentiTFIDF – Sentiment Classification using Relative Term Frequency Inverse Document Frequency
Sentiment Classification refers to the computational techniques for classifying whether the sentiments of text are positive or negative. Statistical Techniques based on Term Presence and Term Frequency, using Support Vec...
Edge Detection on DICOM Image using Triangular Norms in Type-2 Fuzzy
In image processing, edge detection is an important venture. Fuzzy logic plays a vital role in image processing to deal with lacking in quality of an image or imprecise in nature. This present study contributes an authen...