Makine Öğrenmesi Algoritmaları Kullanılarak Türkiye ve AB Ülkelerinin CO2 Emisyonlarının Tahmini

Abstract

Especially in the 20th and 21st centuries, the acceleration of the industrial revolution, the global warming of the world, and the increase in the amount of temperature caused the cycles and intensities of the seasons on earth to change. The increase in the use of energy needed in all areas of life, especially the orientation to fossil fuels, has caused and continues to cause an increase in the emission of CO2, which has an important share among greenhouse gases, and climate changes. Therefore, the management, follow-up, and analysis of CO2 emissions caused by human activities have gained importance for all countries in the world. This study, it is aimed to examine the estimation and trends of CO2, a greenhouse gas associated with climate change, in Turkey and EU countries. In line with this goal, an estimation was made for CO2 emissions that occur when energy is obtained by using decision tree modeling, artificial neural networks, and support vector machines from machine learning algorithms. Success performance of machine learning methods was evaluated using decision tree modeling score value, mean square error (MSE) and root mean square error (RMSE). Support vector machines (SVM) and artificial neural networks (ANN) methods were applied, and estimations were made by using the population, CO2 emission amount, and energy consumption amount for the countries in the years 2010-2019. According to the results, it has been determined that the SVM method gives more successful results than the ANN method. When the European Union countries are considered as a whole, it has been observed that the amount of CO2 emissions have decreased and increased, that is, the rate of increase has decreased. In addition, reductions in CO2 emissions are expected in Germany, the Netherlands, Italy, and France. For Turkey, the situation is different. The amount of carbon dioxide emissions between 2010-2019 in Turkey is increasing every year, and it has been seen in the study that this increase will continue if alternative energy sources that do not emit CO2 are not used in the coming years.