An application study for using fuzzy logic in risk analysis

In order to be able to produce solutions to the problems we face in our daily lives, researchers are trying to transfer the focused processes to analysis systems as realistically as possible. The success of implementation studies depends on the ability to transfer the highest level of information to the analysis process. This has inreased the necessity for verbal variables that increase the wealth of information in the analysis, however make the solution difficult. With more and more usage of linguistic variables due to the use of fuzzy logic based methods, the formation of solution systems similar to human thinking process are becoming widespread. Factors affecting risk and explaining risk levels are predominantly verbal in risk analysis studies. Fuzzy logic based methods provide a transition structure between classes in accordance with the structure of linguistic variables, therefore it is a reason for using fuzzy methods in studies. In this study, a risk analysis study was performed using fuzzy c-means methods from fuzzy clustering methods. Market surveillance data are used as application data. General information about the study is given in the first part, which is the entrance section. In the second and third chapters, the concepts of risk and market surveillance that application data are used, are examined. In the fourth chapter, fuzzy logic and fuzzy set concepts are briefly mentioned, and, information about fuzzy c-means methods used in practice are given in the fifth part. The sixth section contains the details of the implementation and the results of the k-means method, which is a classical logic application with the fuzzy c-means method, are evaluated and the classification success of the methods are compared. In the seventh chapter, which is the final chapter, the results were evaluated and suggestions are made for improvement of the study. Keywords: Fuzzy Clustering, Fuzzy C-Means, Market Surveillance, Risk Analysis.