@article{Kaewtapee_Khetchaturat_Nukreaw_Krutthai_Chaiyapoom Bunchasak_2021, title={Artificial neural networks approach for predicting methionine requirement in broiler chickens}, volume={51}, url={https://he01.tci-thaijo.org/index.php/tjvm/article/view/247456}, abstractNote={<p><span class="fontstyle0">The objective of this research was to apply artificial neural networks (ANNs) for predicting the methionine<br>requirement in broiler chickens at day 1-10 (starter period) and day 11-21 (grower period). A total of 28 data was<br>obtained from five hundred and sixty male broiler chicks (Ross 308), which were divided into twenty-eight pens with<br>twenty chickens in each. Body weight was determined at days 10 and 21. A bootstrapping method was used to multiply<br>the observations to overcome the limited data for training. A total of 280 data was obtained and divided into a training<br>set (</span><span class="fontstyle2">n </span><span class="fontstyle0">= 220) and a testing set (</span><span class="fontstyle2">n </span><span class="fontstyle0">= 60). The level of TSAA supplementation (%) was used as a variable in the input node,<br>whereas the average daily gain (g) was used as a variable in the output node. The model evaluation was determined<br>by R</span><span class="fontstyle0">2</span><span class="fontstyle0">, mean absolute deviation (MAD), mean absolute percentage error (MAPE) and mean square error (MSE).<br>Quadratic regression and ANNs with radial basis function were used to develop the model using Python programing.<br>The results showed that no significant difference (</span><span class="fontstyle2">P</span><span class="fontstyle0">>0.05) was observed in means between the original data and the<br>bootstrapping data. The ANNs showed greater accuracy of R</span><span class="fontstyle0">2 </span><span class="fontstyle0">when compared with quadratic regression at the starter<br>(0.7178 </span><span class="fontstyle2">vs</span><span class="fontstyle0">. 0.7294) and grower (0.8086 </span><span class="fontstyle2">vs</span><span class="fontstyle0">. 0.8097) periods. For error measurements, ANNs also resulted in lower MAD,<br>MAPE and MSE when compared with quadratic regression at the starter and grower periods. In conclusion, the optimal<br>level of methionine (as total sulphur amino acids) obtained by ANNs was 1.13 and 0.99% for starter and grower periods,<br>respectively. Therefore, ANNs are an alternative method to predict methionine requirements of broiler chickens for<br>improving poultry production.</span> </p>}, number={1}, journal={The Thai Journal of Veterinary Medicine}, author={Kaewtapee, Chanwit and Khetchaturat, Charn and Nukreaw, Rattana and Krutthai, Nuttawut and Chaiyapoom Bunchasak}, year={2021}, month={Jan.}, pages={161-168} }