Optimization of student learning status by instructional intervention decision-making techniques incorporating reinforcement learning

Abstract The aim of this study is to solve the problem that existing instructional intervention decision-making techniques are difficult to achieve accurate interventions when optimizing students’ learning contexts. Therefore, this study combines the reinforcement learning model and the quantile tracking regression model to construct a theoretical model for instructional intervention decision-making and validate its effectiveness. The experimental results showed that the proposed model had high prediction accuracy in different student groups, and its application in practical teaching practice could obviously improve students’ learning effectiveness. Compared with the comparison method, the research model performed better in accuracy, precision, recall rate, and F 1 value, and the accuracy was as high as 96.4%. In different educational data sets, the F 1 scores of the proposed model were all above 0.89. The results show that the model can achieve accurate teaching intervention, thus optimizing students’ learning conditions. The research lays the foundation for creating a more intelligent and adaptable educational system, and promotes the intelligent development of educational technology.