Student Success Prediction Using Feedforward Neural Networks

Machine learning algorithms have been used in the last decade to predict human behavior. In education, the student's behavior, and accordingly, their success prediction is also applicable in parallel with the developments in machine learning algorithms and the increased availability of the datasets. The datasets include the observations, which the machine can learn to predict student behavior. By this analysis, given the background information about a student, the features representing a student sample, and the student's possible performance may be estimated. This study's motivation is to predict a student's possible performance to give guiding service. This paper proposes a novel approach for predicting student success by using conventional feed-forward neural networks. The algorithm selects the most informative features based on the variances and uses those features to represent a student sample. The approach is tested on the Experience-API (X-API) dataset collected from Kalboard 360 e-learning system. There are 480 samples in total, with 16 features. It is shown that the improved approach achieves comparable results around 91.95% acceptable predictions by only using behavioral attributes and 93.17% acceptable prediction rates without the feature selection process, respectively.