Design and Validation of an Engineering Management Learning Path Optimization Model Based on Intelligent Recommendation

Mengjiao Zhou
Chongqing College of Mobile Communication, China

Abstract:

With the continuous complexity of the engineering management course system, the traditional fixed teaching path has been difficult to meet the diverse and dynamic learning needs of students, and there is an urgent need to build an efficient and personalized learning path recommendation mechanism. In this paper, based on knowledge graph construction, user portrait modelling and deep learning recommendation algorithm, a learning path recommendation system integrating reinforcement learning optimization strategy is designed. The system realizes structured modelling of learner ability through graph attention network, combines two-tower neural network and multi-objective loss function for learning path ranking prediction, and realizes adaptive optimization of paths through Actor-Critic structure in dynamic environment. In the comparison experiments with collaborative filtering, GCNRec and LSTM path recommendation models, the proposed system significantly outperforms the benchmark model in multiple dimensions such as Precision@5, NDCG@5 and path coherence score. The results show that this method can effectively improve the recommendation accuracy, interest fit and ability suitability, and provides a generalizable path optimization technical solution for engineering education intelligence.