An Evaluation Method for University Classroom Education Quality under Machine Vision and Single-Valued Neutrosophic Hesitant Fuzzy Set Environment

With the advancement of artificial intelligence, machine vision offers a novel approach to university teaching quality evaluation (TQE). However, existing studies are often hindered by subjectivity and lack of standardized evaluation methods, which impede accurate assessment of student learning effectiveness. Therefore, this study addresses these limitations by proposing a TQE framework that integrates machine vision with single-valued neutrosophic hesitant fuzzy sets (SVNHFSs). Specifically, the main contributions of this study are as follows. First, this study innovatively employs machine vision to capture student learning behaviors, constructing a classroom behavior matrix that serves as the foundation for evaluation. Second, this study introduces a combined weighting method, leveraging both the entropy weight method and the Criteria Importance Through Inter-Criteria Correlation (CRITIC) weight method, to assign weights to different time-points during the classes. Third, the SVNHFS is utilized to construct a classroom behavior evaluation matrix, and the single-valued neutrosophic hesitant fuzzy weighted average (SVNHFWA) operator is applied for weighting. In addition, the cosine measure is employed to rank time-points based on both ideal and non-ideal solutions, obtaining the optimal and non-optimal learning effectiveness periods. Finally, a case study confirms the effectiveness and feasibility of the proposed model, offering a robust method for evaluating university education quality.