Reliability Evaluation of Ideological and Political Teaching Effect in Colleges and Universities Based on Bayesian Inference

Due to the influence of different reliability dimensions on the reliability evaluation of teaching effectiveness, the evaluation results are not accurate. Therefore, a Bayesian inference based method for evaluating the reliability of ideological and political teaching effectiveness in universities is proposed. Calculate the posterior probability of Bayesian inference, and extract prior information from existing data through learning based on the learning results of Bayesian network sample data. Through the reasoning stage, the direction dimension, theoretical dimension, transformation dimension, and service dimension of the reliability of teaching effectiveness are evaluated through reasoning. Based on the posterior information, infer the results and calculate the multidimensional integral value. The use of equidistant method to calculate data discretization information entropy reflects the uncertainty of the sample set. Construct a reliability evaluation model for teaching effectiveness, combined with Bayesian hierarchical regression model, to assess the reliability of ideological and political teaching effectiveness in universities. The experimental results show that there is a one point error between the lowest score analysis result and the actual data using this method, while the highest score analysis result is consistent with the actual data, indicating that the evaluation results using this method are accurate.