<![CDATA[E-Learning is a commonly used platform by most institutions, especially during the pandemic Covid-19. E-learning services include viewing, submitting, and uploading files, attempting quizzes, viewing forums, and downloading files. The data store in the servers grow on par with the increment of users in e-Learning@UTM every semester. As a result, the data have become extremely huge. These web log data can be used in augmented analytics to find meaningful insights. The web log data extracted are the log files of the history engagement of users and students’ grades. Data obtained are used in augmented analytics to study the pattern of the data and insights into meaningful information. This research focuses on classification of data through predictive analytics. Hence, predictive models are required. To prove a better outcome, building the model consists of three types of algorithms; Decision Tree, Artificial Neural Networks and Support Vector Machine which are used and compared. After extracting data from e-learning, the first step in building a predictive model is to do data collection, data pre-processing, and data transformation. These three classifiers use the pre-processed data and split the data into training and test sets afterwards. Each classifiers techniques are built and a confusion matrix is applied as a performance measurement to summarise the performance of a classification algorithm, respectively.]]>
Copyrights © 2023
