Evaluation of financial statements by using clustering analysis which is one of data mining techniques in project supports


Thesis Type: Postgraduate

Institution Of The Thesis: Bursa Uludağ University, Sosyal Bilimler Enstitüsü, Turkey

Approval Date: 2019

Thesis Language: Turkish

Student: Şevket Candaş

Supervisor: AZİZE GÜL EMEL

Abstract:

The aim of the study is to create a new method that enables the financial analysis of a large number of enterprises applying to project offer calls. The data set used in the study has been obtained from the balance sheet and income statements of SMEs in different sectors and scales. Enterprises are grouped according to their variable values by using clustering method, one of the data mining techniques. Numerous variables have been obtained through vertical analysis from the balance sheets and income statements of the enterprises. The analyzed data set consists of 10 separate attribute values of 199 enterprises. The enterprises in the data set were analyzed according to the financial characteristics of the clusters and the findings were interpreted. In line with the information obtained, financially appropriate clusters were identified and it was proposed to make positive discrimination in the project evaluation processes for the enterprises in these clusters. The clusters obtained by the applied method and financial differences were found to be significant. As a result, it has been understood that the method applied enables the enterprises to be compared financially and it can be used for the purpose determined in the study.