Research on the Construction and Decision Optimization of Dynamic Evaluation System for Accounting Education Quality Based on Apriori Algorithm

The deep integration of artificial intelligence and big data technology puts forward higher requirements for the dynamic adaptability of education management system. Aiming at the problems such as weak curriculum relevance and lagging management decisions in the quality assessment of accounting education, this study constructs a dynamic assessment system for the quality of accounting education based on the Apriori algorithm and multi-source data-driven. By integrating 2.1TB of daily structured and unstructured data (including accounting practice operation logs, course grades, and case teaching videos), the system realizes three core functions: course group association rule mining (support level 0.09, confidence level 0.62), learner 32-dimensional ability portrait modeling and resource accurate recommendation (match level 82%). The empirical results show that the system can improve the efficiency of management decision making by 58%, and the satisfaction of students' personalized service is 91%, and the time series analysis validates the significant promoting effect of data structure courses on professional competence cultivation (Pearson coefficient 0.62, p<0.01). The research breaks through the static assessment framework and provides a closed-loop solution of “monitoring - early warning - optimization” for Accounting Education management, but it needs to further explore multi-modal data fusion and regional collaboration mechanisms.