Research on the Application of a High-dimensional Bayesian Network in the Evaluation of Ethnic-Chinese Bilingual Teaching

There is still an imbalance in the quality of basic education in ethnic minority areas, especially in the area of ethnic-Chinese bilingual education resources. Developing ethnic-Chinese bilingual teaching is a task that must be undertaken as a matter of priority. Carrying out feasible evaluation of bilingual teaching is one of the necessary conditions for standardizing and leading the development of bilingual teaching. In assessing the problems in the evaluation of bilingual teaching in the ethnic-Chinese language, there are many indicators, intricate relationships, and regional differences. This paper proposes a performance evaluation model of bilingual teaching based on high-dimensional Bayesian network to find out the causal relationship between the weak indicators of input and output in bilingual teaching. The experiments show that through basic data processing, causality measurement, reduction of clustering, and fusion of hill-climbing algorithms for causal relationship scoring to construct subnets and keyword networking, a high-level Bayesian network bilingual teaching performance evaluation model can accurately analyze the causal relationship between weak indicators to provide decision support for rational allocation of resources and improvement of teaching output.