To address the current bias and subjectivity in online teaching quality evaluation in higher vocational colleges, this paper proposes an online teaching quality management method based on data mining algorithms. The Apriori algorithm analyzes teaching data, calculates the support and confidence levels of the data, and adaptively determines the threshold using high-order polynomial curve fitting. Next, the paper systematically identifies key factors influencing teaching quality from three dimensions: the teaching platform, teachers, and students. A quantitative indicator system is constructed, including student satisfaction and platform stability. Based on this, an online teaching quality management framework is designed, comprising a driving layer, an action layer, a guarantee layer, and an evaluation layer, supporting functions such as grade processing, entry, and query. Grades are assessed using a parallel grading and percentage system, achieving scientific grading through discretization and critical value division. Experiments show that, with 400MB of teaching resources, this method achieved a support value of 0.22 and a confidence level of 0.72, demonstrating high student usage. After applying this method, the download and usage rate of core course supplementary materials reached 85%, and the success rate of video-assisted learning in practical courses increased from 70% to 88%. In an evaluation of five teachers, this method’s error rate was generally lower than that of the comparison method, validating its effectiveness and superiority in accurately assessing teaching quality and supporting management decision-making.
