Research on Automatic Analysis of College Students' Ideological Learning Material Demand Based on LSTM

With the continuous development of personalized education, how to effectively analyze and predict students’ demand for learning materials has become an important problem to be solved in the field of education. Especially in the ideological and political courses of college students, how to accurately predict their demand for learning materials based on their learning behavior data is still a challenge. Traditional data analysis methods mostly rely on static features and simplified models, which are difficult to capture the temporal and complexity of student demand changes. This study proposes an automated demand analysis method based on long short-term memory (LSTM) network to solve this problem. This study collects and preprocesses students’ learning behavior data, extracts features such as learning time, learning frequency, and participation, and constructs an LSTM model. The model can effectively capture the temporal characteristics of students’ learning needs and predict future learning material needs through historical data. In the experiment, the model showed high prediction accuracy and can accurately reveal the demand patterns of different student groups in ideological and political courses. The experimental results show that the LSTM model is significantly better than traditional data analysis methods in predicting students’ demand for ideological and political learning materials. Through the optimized model, the prediction error is gradually reduced, and the prediction accuracy is improved by about 15%.