<![CDATA[This study aims to evaluate the impact of social media usage on the academic productivity of Universitas Sanata Dharma Yogyakarta students, measured through their Grade Point Average (GPA). The methods employed involve two machine learning models: Decision Tree and Random Forest. The data were processed using outlier-resistant scaling techniques and data balancing through oversampling. The results show that the Random Forest model outperformed with an accuracy, precision, recall, and F1-score of 90% each. Meanwhile, the Decision Tree model achieved 80% accuracy, with a precision of 86%, recall of 80%, and F1-score of 82%. Feature importance analysis revealed that 'Faculty' and 'Gender' are the most significant factors in predicting students' GPA. This study concludes that employing Random Forest with data balancing techniques can enhance prediction accuracy and reliability, providing insights into the optimal use of social media to improve students' academic productivity.]]>
Copyrights © 2024
