Hybrid dynamic clustering algorithm in audit sampling

作者：Feng Yinbo^*

来源：International Journal of Applied Mathematics and Statistics, 2013, 51(23): 62-69.

• Clustering number
• Dynamic clustering
• Hybrid dynamics
• K-means
• k-Means algorithm

Clustering can find the potential rules and relations in data better, and if it is applied to the study of audit risk management, audit ground and judgment can be offered to the auditors. In order to solve this problem, this paper has studied audit sampling and proposed hybrid dynamic clustering algorithm model based on K-means and CURE, which centrally clusters the sensibility of original values and the selected objects rapidly by K-means algorithm, then eliminates isolated points in the model by CURE algorithm, and last determines the clustering numbers. The simulation experiments show that the performances of hybrid dynamic clustering algorithm model proposed in this paper are much better than the common K-means and CURE algorithm.

• 出版日期2013
• 单位河南大学