The Academic Warning Model Based on TLBO-BP Algorithm

Abstract

Tracking students' learning situations, analyzing test scores and predicting possible states of learning in the future are helpful for students who need academic intervention. And it is an important task with practical meaning and research value for teachers to provide targeted guidance and a decision-making basis. In this paper, a group intelligent optimization algorithm and the neural network theory are applied to the students' academic intervention problem. The TLBO-BP algorithm is proposed to optimize the initial weights in the BP neural network. It can speed up the convergence of the algorithm and avoid the instability of the training result caused by the random initialization of the BP neural network. The model shows satisfactory results in two examples, which further supports its validity.