Optimisation of network learning platform based on machine learning algorithm
Jin Zhang · International Journal of Computer Applications in Technology · 2025
The purpose of this study is to optimise online learning platforms through deep learning and address issues related to personalised user experience and resource allocation. A comprehensive optimisation framework is proposed, comprising three modules: user behaviour analysis, personalised recommendation and resource optimisation scheduling. First, a recommendation mechanism is developed by integrating Neural Collaborative Filtering (NCF), the Transformer model and Content-Based Filtering (CBF) techniques. Accordingly, a user behaviour prediction and personalised recommendation model based on a fused NCF-CBF-Transformer algorithm (NCF-CBF-T) is constructed. This model enhances the personalised recommendation system by leveraging multi-level technology integration. Specifically, the Transformer model captures temporal dependencies in user behaviour sequences and dynamically models long-term user interest evolution through the multi-head self-attention mechanism. This study contributes to the theoretical advancement of deep learning applications in educational technology and provides practical experimental references for optimising online learning platforms.