Personalized Online Ideological and Political Education Services Based on Deep Learning

As the big data and other technologies develop, personalized online education services have become the focus of general attention. This paper focuses on the learning behavior data of students in online ideological and political education, and based on the latest deep learning and education data mining theories, provides better technical solutions for personalized online ideological and political education services. It proposes a gated loop network, which divides user behavior sequence into multiple historical sessions as input, and then uses attention mechanism to extract the user's interest in each session, the user's dynamic preference and the user's short-term preference respectively, and finally obtains the user's mixed preference representation. The experimental results show that the recommendation performance of this recommendation model is better than that of the benchmark model.