Prediction of FL 305 DMY from monthly part lactation milk yield records using artificial intelligence in Sahiwal cattle

Abstract

The present investigation was carried out on 4,760 first lactation monthly part lactations milk yield records of 476 Sahiwal cows sired by 34 bulls maintained at National Dairy Research Institute (NDRI), Karnal over a period of 51 years (1961-2011) to predict first lactation 305-day milk yield (FL305DMY). The comparison was made between multiple linear regression analysis and artificial neural network method for prediction of first lactation 305-day milk yield in Sahiwal cows. In multiple regression method, an optimum equation having 3 variables namely PL2, PL5 and PL8 gave an accuracy of prediction of 88.80%, whereas under artificial neural network, this equation gave an accuracy of prediction of 89.29%. It was concluded that artificial neural network could be used as an alternative to multiple regression method for prediction of FL305DMY in Sahiwal cows.