A knowledge tracing model based on attention mechanism

For the past few years, with the gradual promotion and application of personalized online education system, researchers began to consider using online education system to trace and analyze students' knowledge status, so as to teach students according to their aptitude and improve their learning ability. At present, researchers use the deep knowledge tracing method to research students' academic performance based on the interaction records of students' answers and build deep knowledge tracing model, which has shown good performance. However, the deep knowledge tracing model does not carry out in-depth analysis of knowledge interaction, leading to its poor interpretability. Therefore, in this study, we regard attention mechanism as an effective interpretability module to construct a new knowledge tracing model, which effectively improves the interpretability and predictive ability. The experimental results also indicate that the expression of this model is improved to a certain extent.