Research on predicting learning achievement in a flipped classroom based on MOOCs by big data analysis. Issue 1 (10th September 2021)

• Record Type: Journal Article
• Authors: Qian, Yan; Li, Cheng‐Xin; Zou, Xiu‐Guo; Feng, Xue‐Bin; Xiao, Mao‐Hua; Ding, Yong‐Qian
• Is Part Of: Computer applications in engineering education. Volume 30:Issue 1(2022)
• Page Start: 222
• Page End: 234
• Publication Date: 2021-09-10
• Subjects: artificial neural network -- educational data mining -- flipped classroom -- MOOCs -- online education; Engineering -- Study and teaching (Higher) -- Periodicals; Engineering -- Computer-assisted instruction -- Periodicals; 620
• DOI: 10.1002/cae.22452
• Languages: English
• ISSNs: 1061-3773

Abstract

Abstract: In the era of big data mining, educational data mining has become a principal research focus, with online education mining, such as massive open online courses' (MOOC) data analysis, representing an important source of it. Recent studies have found that learners have low passing rates on MOOCs. A number of studies have proposed prediction models for the dropout rate of learners on MOOCs. The improvement of MOOCs and the promotion of personalized education are the key points of online education. However, the selection and intervention of students with a tendency to drop out slows down the efficiency of teaching and increases the burden on teachers. This study's aim is to utilize back propagation neural networks and radar graphs in a flipped classroom based on MOOCs to predict students' future grades and to analyze the influence of teaching from various perspectives to support the promotion and reform of teaching and curriculum. Compared with the previous year, after the forecast and the adjustment, this year's student scores increase significantly.