Abstract
Objective: Education acts as a significant role in student’s life were low scholastic performance create a vast impact on the final level of the scholars. However, no model will guide the student and teachers in predicting academic performance and subsequently help improve the ranks of the students. Methods: Data mining technique explicitly utilized the J48 algorithm to predict the academic performance of the students. The 10-Folds Cross-validation and Receiving Operating Characteristics Curve (ROC) was deployed to create a model and test the result based on the attributes. The collected datasets of this study are from the previous grades of the 2nd year BSIT students enrolled in the Computer Organization Course from S.Y. 2016-2017 and 2017-2018. Findings: The result generated in the decision tree model and decision rule classification, Confusion matrix, ROC and AOC show that Lab exercise/Project is the most critical attribute that profoundly affects the students’ academic performance followed by quizzes, finals, recitation and midterm attribute in the Computer Organization class. Additionally, from the result, the model was able to identify students who will pass at 89.0% accuracy, failed at 92.60% accuracy and conditional at 74.90%. Finally, the model has high acceptability and accuracy rate in predicting the Student Academic Performance in Computer Organization. Application/Improvements: This study can be used to develop or create a model that will predict the academic performance of the students in Computer Organization. For more improvement of the subject area, it recommended other data mining technique would be to predict academic performance with additional parameters to test the accuracy of the algorithm. Keywords: Academic Performance, Computer Organization, Data Mining, Decision Tree, Information Technology
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have