Arbitration of Turkish agricultural policy impact on CO2 emission levels using neural networks

The relationship between environmental pollution and economic growth has lately been the focus of discussion among policy makers and scientists worldwide. In Turkey, the agriculture sector is a major contributor to the national economy and consequently growth in agriculture leads to economic growth; which in turn affects pollution levels. In this paper, we present a novel application of neural networks based on using the simple and yet efficient back propagation learning algorithm. We estimate the carbon dioxide (CO2) emission levels by approximating the non-linear relationship between agricultural factors and alterations in CO2 emission levels. We use a public Turkish dataset spanning 1968-2010 and showing recorded major agricultural and economical indictors and the level of CO2. The experimental results indicate that such approximation is successfully possible and encouraging to be used in more similar non-linear applications. (C) 2016 The Authors. Published by Elsevier B.V.