ACADSTAFF UGM
| Title | : | Prediction of Final Score Tendency in e-Learning Based on Classification of Log Data |
| Author | : |
Isna Alfi Bustoni, S.T., M.Eng. (1) |
| Date | : | 7 2020 |
| Keyword | : | Classififcation, e-Learning Classififcation, e-Learning |
| Abstract | : | Gamification is the implementation of a slice or a whole game design concept into a real problem. This concept has been adapted to e-learning massively, as well as in the Massive Open Online Course (MOOC). In the e-Learning, system stores learning activity data for each user, namely the log data. This log data can be used to conduct a quantitative evaluation of an information system, such as prediction and classification. In some studies student activity can be used to predict the tendency of student grades. A commonly used method for recent years is Support Vector Machine (SVM). SVM is a well known method that able to show good performance for classification, especially for a small number of classes. Meanwhile Random Forest is known as a method that is able to manage data with a large number of features. However, there are still limited research that compare both of this method in order to predict the tendency of student final score at the end of e-Learning. Thus, our study aim to classify activity log data to predict student final grades tendency using both Random Forest and SVM as well as compare their performance. |
| Group of Knowledge | : | |
| Level | : | Internasional |
| Status | : |
Submitted
|
| No | Title | Action |
|---|