Comparison and prediction with Gompertz growth curve and artifical Neural Networks Models in broiler chickens

In this study, growth curves were estimated and compared with Gompertz model and ANN (Artificial Neural Network) which is most commonly used in broiler chickens in solving nonlinear problems. In comparison of methods; determination coefficient (R²), mean absolute deviation (MAD), mean absolute percent error (MAPE), and mean squares error (MSE) statistics were used. According to the results obtained; the method with the highest R² value was found to be the best model among the studied models. In the study obtained, MSE values were found to be 5625 and 2950 for ANN and Gompertz model, MAPE values were 0.27 and 0.17, and MAD values were 0.5 and 1.2, respectively. The MAD value determined by ANN model was found to be lower. R² values were determined 0.98 and 0.99 for ANN and Gompertz model, respectively. It was determined that the R² values obtained from ANN and Gompertz models and MAPE values are very close to each other. It has been shown that MSE value in the Gompertz model gives better results than ANN. As a result; the ANN model is thought to be successful in explaining growth and can be used as an alternative model to the Gompertz model, which is one of the best models in this respect.