Reform and Effectiveness Assessment of Accounting Teaching in Colleges and Universities Based on Multi-source Data Mining

Abstract

The growing discrepancy between the supply and demand of new accounting professionals underscores an urgent need for reform in the educational paradigms within higher education accounting programs. This paper introduces a comprehensive proposal for educational reform in accounting at the collegiate level, using ‘A school’ as a pilot site for our experimental teaching reform. Central to our methodology is the design of a multi-source data mining process, utilizing characteristics of educational information data. Employing the K-means clustering algorithm as a foundation, we further develop a Fuzzy C-Means (FCM) clustering algorithm to evaluate the impact of these pedagogical reforms. Subsequent to the reform, we collected and analyzed teaching data, establishing specific assessment metrics. The analysis, conducted via clustering, revealed significant improvements. Prior to the reform, the experimental class scored an average of 62.89, compared to the control class’s 63.05. Post-reform, the experimental class’s performance elevated to 74.57, while the control class saw a more modest increase to 65.14. This marked enhancement in the experimental class’s performance underscores the efficacy of the reform, aligning with our educational objectives, which anticipate an 80% qualification rate and a 42% excellence rate among accounting majors. This study not only advances the theoretical framework but also refines the practical approaches to cultivating accounting talent in collegiate settings. It offers valuable insights and serves as a reference for future reforms in accounting education at universities.

Keywords

• multi-source data mining
• k-means algorithm
• fcm clustering algorithm
• assessment index
• accounting teaching reform
• 68t05