Language Personalized Recommendation Algorithm Based on NCF
Zhuo Wang, Rui Tang, Si Wu · 2024
Traditional algorithms based on keyword matching and simple statistical methods for recommendation cannot effectively understand semantic and contextual information, resulting in limited effectiveness in handling complex semantic relationships and user interest modeling. This study adopted the NCF (Neural Collaborative Filtering) algorithm, which combined collaborative filtering with neural networks. By analyzing users' historical behavior and learning preferences, a user project interaction model and a learning material feature model were constructed. Then, the neural network model was used to learn and utilize the idea of collaborative filtering to predict. The results showed that the accuracy of the proposed method was as high as 96%, and the NCF based language personalized recommendation algorithm achieved significant results in language learning resource recommendation. This algorithm can improve recommendation accuracy by accurately predicting users' preferences for learning resources and recommending learning materials that meet their personalized needs.