Decision Tree for 305-Day Milk Yield in Cross-bred Cattle
Rohit Sharma
Lala Lajpat Rai University of Veterinary and Animal Sciences, Hisar, Haryana, 125004, India.
J. K. Chaudhary
Department of Animal Genetics and Breeding, CVSc. and A.H., Central Agricultural University (CAU), Aizawl, Mizoram-796014, India.
Renuka Hada *
College of Veterinary and Animal Science, RAJUVAS, Navania, Udaipur, Rajasthan, 313601, India.
Parth Gaur
National Dairy Research Institute (ICAR-NDRI), Karnal, Haryana, 132001, India.
Smriti Sharma
Lala Lajpat Rai University of Veterinary and Animal Sciences, Hisar, Haryana, 125004, India.
N. Shyamsana Singh
Department of Animal Genetics and Breeding, CVSc. and A.H., Central Agricultural University (CAU), Aizawl, Mizoram-796014, India.
T. C. Tolenkhomba
Department of Animal Genetics and Breeding, CVSc. and A.H., Central Agricultural University (CAU), Aizawl, Mizoram-796014, India.
*Author to whom correspondence should be addressed.
Abstract
Cows' Lactation Milk Yield (LMY) is a crucial factor in animal breeding operations. Investigating the influence of potential environmental factors on lactation milk yield is of paramount importance and in order to identify the various factors influencing lactation milk yield, dairy cattle records were analysed using the regression tree approach. Age, Parity (P), Lactation Length (LL), and Calving Season (CS) were taken into account as explanatory variables while 305-day Milk Yield (MY) as a dependent variable. Decision tree study revealed that Lactation Length, followed by Parity, Age, and Calving Season, had the greatest impact on the 305-d milk output of cross-bred cows. The regression trees use the tree to represent the recursive part. Each terminal node or leaf of the tree represents a cell of the section and has just added a simple pattern applied to it in this cell. It was evident from nodes (branches) in regression tree, that cows with parities of 1 and 4 (node 11) produced less milk than cows with parities of 2, 3, 5, 6, 7 and 8 (node 10). More milk was produced by cows older than 4.3 years and whose calving seasons were spring and summer (node 40). With the use of the regression tree method, we were able to extract sub-homogenous groups based on the explanatory variables from records of cross-bred cattle and determine the combinations of environmental conditions that produced the maximum 305-d milk yield.
Keywords: Lactation length, parity, milk yield, decision tree, regression tree