Recently, flipped learning has been shown that it is better than traditional instruction by in-class lecture mode. An important issue for flipped learning is to collect and analyze students’ learning portfolio before class so that teachers can obtain their learning situation as soon as possible before lecturing. However, existing MOOCs systems do not provide functions performing to collect and analyze students’ learning portfolios. Therefore, the thesis proposes a MOOCs system called the gCourse system. One main feature of the system is to quickly collect students’ portfolios and then analyze their portfolios. As a result, teachers utilize the gCourse system to refer to the results before class. These results can help teachers to adjust their curriculum plans or pedagogical methods during lecturing if necessary. These results also assist teachers in planning out-of-class activities for students.