Investigating the Effect of Different Data Clustering Methods on the Accuracy of Models Related to Accounting Estimates by Comparing Traditional and Classical Clustering Methods

Today, the use of accounting information estimation is the same as other disciplines because of the lack of access to all information. For this reason, in this research, we tried to study the accuracy of accounting estimation models using different clustering methods to determine how different clustering methods increase the accuracy of the desired models and the preferred method Among the different clustering methods, which method can be used to increase the accuracy of the models. The research sample consisted of 99 companies listed in the Tehran Stock Exchange. In order to collect the required data, the financial statements and notes of the 9-year period (2008-2017) were used by the companies. The results of the research showed that the use of different clustering methods increases the accuracy of accounting estimates models in most cases. However, among the clustering methods used in the research, the classic clustering method is a more appropriate method than the method The traditional approach is to increase the accuracy of accounting estimates models.