Affiliations: [a] School of Public Policy and Administration, Nanchang University, Nanchang, Jiangxi, China
| [b] Hohhot No. 2 High School, Hohhot, Inner Mongolia, China
Correspondence:
[*]
Corresponding author. Zhu Yi, School of Public Policy and Administration, Nanchang University, Nanchang 330000, Jiangxi, China. E-mail: zy_15170249712@163.com.
Abstract: Online teaching platforms have developed into mainstream knowledge learning and exchange platform. The research on the quality evaluation of online teaching platforms and the construction of an applicable and scientific evaluation index system model can help explore the key factors affecting the quality of online teaching platforms and provide some references for evaluating online teaching platforms and improving online teaching quality. This study combines the rough set theory (RS) with the BP (Back Propagation) neural networks to build an RS-BP neural network model to evaluate the quality of online teaching platforms. Firstly, an initial online teaching platform quality evaluation index system is constructed based on knowledge transfer theory from four aspects: course content, knowledge transmitter, knowledge receiver and teaching platform. Then, 12 core evaluation indicators were generated by attribute reduction using rough set theory, and the weights of each core indication were determined. The normalized data input was then trained, validated, and tested to generate a rough set neural network quality evaluation model for online teaching platforms. After that, three representative online education platforms of content, interaction and compatibility are selected for empirical research. The accuracy of the model is first tested by the error between the simulated and output values, after which the core metric scores and the overall scores are calculated for the three types of platforms. The empirical results demonstrate that the model has certain advantages in terms of index simplification and adaptive training when evaluating online teaching platforms, as well as strong operability and practicality. The evaluation results show that the content online teaching platform has the highest comprehensive score, followed by the compatible and interactive online teaching platforms. According to the index scores, the quality of the course content, stage assessments, and contact between professors and students were identified as major elements influencing the quality of the online teaching platform. Finally, suggestions for optimization for each of the three types of online teaching platforms were made based on the core indicators and their weights, as well as the scores and characteristics of the three types of online teaching platforms, with the goal of improving the quality of online teaching platforms.
Keywords: Knowledge transfer, online teaching platform, rough set theory, BP neural network
