Treatment and Application of “Student-Evaluating-Teaching” in Colleges and Universities Based on tensor decomposition

Abstract

At present, most colleges and universities adopt arithmetic average method and relative evaluation method to deal with the “learning evaluation” score, which indicates the teaching quality and teaching level of teachers. Due to the different subjects (students) of arithmetic average, it is impossible to eliminate the influence of different grades, classes and courses on teachers’ quality evaluation. Although the relative evaluation method can solve the influence of the evaluation subject on the evaluation of teachers’ teaching quality, it can’t solve the influence of other factors such as the degree of curriculum difficulty. In this paper, a teacher evaluation method based on tensor decomposition is proposed. Tensors are used to describe the evaluation of teachers by students in different classes and curricula. By tensor decomposition, the influence factors of students and curricula are removed, and the characteristic matrix of teachers is obtained to evaluate the teaching quality of teachers. Through comparative analysis of experiments, the superiority of teacher evaluation method based on tensor decomposition is verified.