Title: Research on the integration and optimisation of MOOC teaching resources based on deep reinforcement learning

Authors: Kun Jiao; Xin Han; Li Xu; Terry Gao

Addresses: School of Accountancy, Hebei Polytechnic Institute, Shijiazhuang, Hebei, 050000, China ' International Communication Centre, Shijiazhuang Information Engineering Vocational College, Shijiazhuang, Hebei, 050000, China ' Computer Department, Hebei Vocational College of Politics and Law, Shijiazhuang, Hebei, 050000, China ' Counties Manukau District Health Board, Auckland, 1640, New Zealand

Abstract: There are some problems in the merging of English MOOC learning information, such as high packet loss rate and high repetition rate. This paper designs a new method of resource data merging with the help of deep reinforcement learning. The operation mode of English MOOC platform was analysed, and teaching resources for the initial application of MOOC under this platform were collected. The pre-processing of the initial collection of resources is completed through translation, Chinese word segmentation, semantic annotation of word segmentation results and other steps. The features of English MOOC teaching resources are extracted, and the deep enhancement learning algorithm is adopted to optimise. Through comparative experiments, it is found that the packet loss rate of the combined learning information is only 0.28% and the integrated resource repetition rate is only 0.89%.

Keywords: deep reinforcement learning; English teaching; MOOC teaching; integration of teaching resources; resource characteristics; feature weight value.

DOI: 10.1504/IJCEELL.2022.126849

International Journal of Continuing Engineering Education and Life-Long Learning, 2022 Vol.32 No.6, pp.763 - 777

Received: 25 Aug 2020
Accepted: 18 Nov 2020

Published online: 09 Nov 2022 *

Full-text access for editors Full-text access for subscribers Purchase this article Comment on this article