Data-Driven Evaluation of MOOC-Based Blended College English Teaching via Enhanced Neural Networks

Abstract

The rapid proliferation of massive open online courses (MOOCs) presents both opportunities and challenges for traditional higher education. As MOOCs offer scalable, high-quality educational resources, they have the potential to significantly enhance instructional outcomes in university settings. In this context, the online-offline hybrid teaching model has emerged as a promising pedagogical approach, particularly in the domain of college English instruction. However, the effective integration of MOOCs into blended learning frameworks remains a complex and evolving challenge. This study presents a data-driven analysis of MOOC-based hybrid teaching for college English. It first identifies key limitations in current implementations, including issues related to interactivity, learner engagement, and instructional design. To address these challenges, a strategic framework is proposed to optimize the blended teaching process. Furthermore, this work introduces an enhanced back propagation (BP) neural network model to evaluate the effectiveness of hybrid English instruction. The improved model incorporates an additional momentum term (AMT), adaptive learning rate (ALR), and a conjugate gradient (CG) optimization algorithm to overcome the limitations of traditional BP networks. Experimental results demonstrate that the proposed model achieves superior performance in terms of accuracy and F1 score compared to conventional methods such as support vector machines (SVM) and deep belief networks (DBN). These findings validate the effectiveness of the proposed framework and highlight the potential of intelligent evaluation models in advancing MOOC-based blended learning environments.