Evaluation of GMDH artificial neural network model to predict the spatial distribution of Coccinella septempunctata (Col.: Coccinellidae) in the alfalfa farm of Bajgah, Shiraz

This study aimed to predict population of Coccinella septempunctata in the field using artificial neural network. The data was collected from a four-hectare field in years of 2013-2014 in the area of Badjga Shiraz. In this model, the input variables were, longitude and latitude and population changes of Coccinella septempunctata was used as the outcome variable. The neural network type used, was Group Method of Data Handling (GMDH) that optimized by genetic algotithm. To evaluate the ability of GMDH neural networks to predict the spatial distribution of the species, statistical comparison of the parameters such as mean, variance, statistical distribution and coefficient determination of linear regression between predicted values and actual values was used. Results showed that in training and test phases of GMDH, there was no significant effect between variance, mean and statistical distribution of actual and predicted values, and the coefficient of determination of 0.98 indicates the high accuracy of this neural network in predicting the density of this species. The drawn maps showed that the distribution of this natural enemy is patchy.