Mining Web-based Learning System Data To Detect Different Pattern Of The Student During Completing Course

The ability to predict student performance,engagement are two important research topics because they can help teachers prevent students from dropping out before final exams and identify students that need additional assistance.The objective of this study is to predict the difficulties and engagement that students will encounter in an e-learning course.We analyzed the data logged by technology-enhanced learning(TEL)systems called digital electronics education and design suite(Deeds)and virtual learning environment(VLE)using machine learning(ML)algorithms.The Deeds system allows students to solve digital design exercises with different levels of difficulty while logging input data.VLE delivers different lectures,assignments,and materials from the Open University(OU)to the students.We then trained the ML algorithms on the data from the training data and tested the algorithms on test data.We performed k-fold cross-validation and computed the receiver operating characteristic and root-mean-square error,recall,kappa and accuracy metrics to evaluate the model's performance.The results show that artificial neural network(ANN)and support vector machine(SVM)achieved higher accuracy to predict student difficulties in e-learning course compared with the accuracy of the other algorithms.Moreover,the results demonstrated that decision trees(DT),J48,JRIP,and gradient boosted trees(GBT)classifiers exhibited better performance in predict student engagement during the VLE course.ANNs,SVM,DT,GBT,and JRIP can easily be integrated into the e-learning system;thus,we would expect instructors to report improved students' performance during the course.