Experimental Demonstration for Learning Analytics and Data Mining on e-assessment

Every subject and course has a learning goal that we work towards during the teaching-learning process, whether it be online or offline. Under the current system, assignments, project analysis, semester exams, and internal exams are used to evaluate students’ learning. Several factors, including participation in class, prior knowledge, engagement time, activity logs, forum posts, understanding level, etc., influence how well students learn throughout a course or session. All these factors will be difficult to analyze in offline mode. All of these activities determine whether or not the learning objective is met. This study examines an experiment that provided blended learning data on students’ academic details, learning behaviors, and evaluation modules to determine the success of the two subjects. It discussed the details of each phase involved in the experiment. Students enrolled in integrated courses, MTech(PG-1) and MCA(PG-2) were analyzed in this study procedure through 4 weeks of offline classes. Moodle Gnomio was used for the online processes related to the Registration process, Quizzes, and discussion forums. A total of 6 quizzes for PG-1(M-Tech) and PG-2(MCA) were organized, and learning resources were provided before the exam. This study discusses detailed experiment setup, and Statistical analysis involving difficulty level of quizzes. This study proposed a framework for the dashboard that utilize this dataset and involves analysis related to student performance, grade distribution, inferring learning patterns, and course success.