Journal of Animal and Plant Sciences
J. Anim. Plant Sci. [ISSN 2071 - 7024]
Volume 4 (3): 373 - 378. Published October 18, 2009.
Estimation of pesticides usage in the agricultural sector in Turkey using Artificial Neural Network (ANN)
*Özbek F. Ş. * and Fidan H.
Department of Agricultural Economics, University of Ankara, 06110 Dışkapı Ankara, Turkey
* Corresponding author e-mail: e113033@yahoo.com
SUMMARY
This study aimed to estimate the amount of pesticides used in the Turkish agricultural sector and to test the hypothesis that pesticides usage will increase in the future. Artificial neural network (ANN) is a data processing technology whose development was inspired by the data processing technology of the human brain, and is used to estimate pesticides usage. In this study, a model has been developed for the estimation of pesticides usage. The feed forward neural network with back propagation model and the radial bases model are the two model types most commonly used in ANN applications. The most appropriate results of these models were selected by comparing ANN model results with actual results. Comparison of the results showed that the the radial bases ANN model is the most appropriate for estimation of the amount of pesticides used in Turkey. It was observed in the forward estimations that pesticides usage increased in 2007 while the amounts used decreased in 2008 (based on usage in 2007). The model predicts increased pesticide usage for the years 2009-2012. .
Key words
Artificial neural network, estimation model, pesticides usage, Turkey
FULL PAPER [PDF AVAILABLE HERE]
