<![CDATA[Basic programming course is one of the courses taken by new students and usually some of those students have difficulties understanding the basic concepts of programming. This study aims to identify students who are struggling on the course at the earliest possible time by using factors that can be collected before any test or evaluation is taken so that the lecturer can provide additional assistance for students who encounter such difficulties. The classification method proposed in this study is Neural Network Backpropagation. Tests will be done to find out whether the proposed method can solve the problem of this study and to find out the best value for parameters such as the number of hidden neurons and hidden layer and learning rate for this study. Some test scenarios are also used in this study such as using all of the data features, using PCA with 85%, 90%, 95% variance, and using only significant features based on Pearson correlation. The test results of this study revealed that the proposed method can be used to solve the problem in this study, with the highest average accuracy of 0.74 in two scenarios, the PCA with 95% variance and using only significant features scenario. Test results also show that the parameters which produce the best result is 7 hidden neurons in one hidden layer and learning rate value of 0.7.]]>
Copyrights © 2019
