Knowledge tracking method based on project reaction principle and GRU-Attention

The invention relates to a knowledge tracking method based on an item reaction principle and GRU-Attention, and belongs to the technical field of deep learning and natural language processing. The method comprises the following steps: firstly, performing deep information feature extraction on an information reading and writing module by utilizing a GRU to obtain a long-term dependency relationship; then, carrying out weight distribution on the feature information through an attention mechanism so as to enhance the attention degree of the model on important information; thirdly, inputting the feature information into a neural network with a hyperbolic tangent function to deduce project difficulty and student ability features; and finally, project difficulty and student ability characteristics are used for project reaction principles to predict correctness of answers. According to the method, the prediction performance and interpretability of the model can be improved, and the prediction performance of the meth