prediction of amino acids contents in corn and wheat by using artificial neural network model and multiple linear regression

to determine the amount of food amino acid and to spend time in the laboratories are expensive & time-consuming due to a chemical analysis. in the current laboratories, digestion nirs method is widely used for this purpose. but this method has technical limitation. therefor is important find appropriate method for estimate amount of amino acids. artificial neural network (ann) can provide a better reflection of the relationship between approximation feed composition and particular nutrient amount in that feed. therefore, this study was performed to estimate amino acids corn and wheat by using artificial neural networks and multiple linear regression (mlr). in neural models used in the study, input variables include crude protein, crude fat, crude fibre, phosphorus and ash, and output variables includ profiles of amino acids relevant to combination of these two types of feed. the results showed that there is a significant relationship between amino acids in corn and wheat and its chemical composition. also the statistical evaluation showed that the ann model compared with mlr was a stronger estimation for prediction the amount of each amino acids. hence the artificial neural network as a powerful tool for modelling, forecasting and estimating the nutrient composition of foods used poultry. using the results of this study, it is recommended that artificial neural network can be used as a computational method with sufficient accuracy for modelling, prediction and estimation of the nutrient composition of foods used in poultry.