College instructors’ evaluation data relies heavily on assessment analysis, yet there is an issue with erroneous assessment. There is an issue with the results from using traditional deep learning to evaluate data from college instructor evaluations. Consequently, this paper’s proposal is to examine college instructors’ assessment data using the association rule method. To begin, we use frequent item set theory to identify the elements that will have an impact, and then we categorize the indicators according to the needs of the assessment and analysis in order to lessen the impact of any interference factors. The outcomes of the evaluation and analysis are then thoroughly examined after an association rule algorithm evaluation and analysis scheme is formed using the frequent item set theory. In comparison to conventional deep learning, the association rule algorithm outperforms it in terms of accuracy and time spent analyzing influencing factors in evaluations, according to the findings of the MATLAB simulations.