Predicting students’ performance in English class

• Published in: AIP conference proceedings Vol. 1977; no. 1
• Main Authors: Purwaningsih, Nunik, Arief, Diana Ross
• Format: Journal Article Conference Proceeding
• Language: English
• Published: Melville American Institute of Physics 26.06.2018
• Subjects: Accuracy; Bayesian analysis; Data mining; Performance prediction; Students
• ISSN: 0094-243X; 1551-7616
• DOI: 10.1063/1.5042876

In the educational realm, data mining is widely used to predict students’ academic achievement. The purpose of this study was to predict the student learning outcomes for English subject. Variable representing a student performance was the final score. Prediction of the final score as the measurement indicator was based on the variables that describe the students profile and their scores on preliminary test. Naïve Bayes classification algorithm was used as the approach to predict the results. The results show that there is a correlation between data composition and accuracy. The higher the percentage of the amount of data, the higher the accuracy of the result. The highest amount of the data was B score (62%), and the highest accuracy was also obtained by the prediction for B score (86%). By doing prediction of students’ performance at the beginning of the class, a lecturer can identify students who need special attention thus they can equal performance with the others.