پيش بيني موفقيت فراگيران دوره‌هاي آموزش الكترونيكي بر اساس ويژگي هاي تحصيلي و ردگيري تعاملات آنها در طول زمان

With the increase in the use of the Internet and communication tools in the world, the use of virtual learning has also increased greatly. But unfortunately, despite the popularity of virtual learning courses, the rate of efficiency and completion of virtual courses is extremely lower than face-to-face courses. By tracking the behavior of learners in the virtual learning environment and analyzing their behavior, patterns can be discovered that can be used to The early prediction of the student's academic status will lead, and in line with that, it is possible to prevent the student's failure at the end of the course by announcing timely warnings to students and other stakeholders, including instructors, course planners, parents, etc. In past studies, by using demographic information and educational record information or using information obtained from user interaction with the virtual education system, an attempt has been made to predict the academic status of students; But based on the studies conducted so far, the investigation of student behavior in the virtual learning environment over time has received less attention. In this research, by focusing on extracting the behavioral characteristics of students in the virtual learning environment, early prediction of students' academic status has been done over time. It is also shown that sometimes paying attention to the behavioral characteristics of students over time from the eva‎luation results obtained by the student during the semester is also more effective for predicting his performance and can be used as a measure to express their success and failure at the end of the course. The results obtained from this research show that after only one third of the semester and having only the first class grade and without having the other three grades of the student (the second grade, the final exam grade and the cumulative grade) with the harmonic mean 93.93% predicted students' success or failure; which is much improved compared to the previous work done on these data.

6