A COMPARISON OF LOGISTIC REGRESSION, ARTIFICIAL NEURAL NETWORKS AND MOORA METHODS IN ESTIMATION OF THE SAFETY OF COUNTRIES Cover Image

ÜLKELERİN GÜVENLİ OLMALARININ TAHMİNİNDE LOJİSTİK REGRESYON, YAPAY SİNİR AĞLARI VE MOORA YÖNTEMLERİNİN KARŞILAŞTIRILMASI
A COMPARISON OF LOGISTIC REGRESSION, ARTIFICIAL NEURAL NETWORKS AND MOORA METHODS IN ESTIMATION OF THE SAFETY OF COUNTRIES

Author(s): Ozlem Deniz Basar, Elif Guneren Genc
Subject(s): Economy, ICT Information and Communications Technologies
Published by: Rating Academy
Keywords: Artificial Neural Network; Logistic Regression Analysis; MOORA; Safety; Classification;

Summary/Abstract: In recent years, because of the developments in software and hardware technology, the datasets used in research have expanded, and with the effects of artificial intelligence technologies, the models used in forecasts have enabled to obtain results with broader meanings. In this study, using the crime index calculated to reveal the crime rates in the countries every year, the safety positions of the 106 countries was estimated. For this purpose, logistic regression analysis, artificial neural networks and MOORA method, which is one of the multi-criteria decision making methods and also not a classification method, has been used to provide a different point of view. As a result of the study, it is determined that the correct classification rate of estimations made according to the safety of countries with artificial neural networks method is higher than other methods.

  • Issue Year: 7/2020
  • Issue No: 2
  • Page Range: 123-134
  • Page Count: 12
  • Language: Turkish
Toggle Accessibility Mode