Comparison between data mining methods to assess calving diffi culty in cattle ¤ Comparación entre métodos de minería de datos para evaluar la difi cultad al parto en ganado
AbstractBackground: Dystocia in cattle results in adverse consequences (increased calf morbidity and mortality, decreased fertility, and milk production, lower cow survival and reduced welfare) leading to considerable economic losses. Objective: To classify calvings in dairy cattle according to their diffi culty using selected data mining methods [classifi cation and regression trees (CART), chi-square automatic interaction detection trees (CHAID) and quick, unbiased, effi cient, statistical trees (QUEST)], and to identify the most signifi cant factors aff ecting calving diffi culty. The results of data mining methods were compared with those of a more traditional generalized linear model (GLM). Methods: A total of 1,342 calving records of Polish HolsteinFriesian black-and-white heifers from four farms were used. Calving diffi culty was divided into three categories (easy, moderate and diffi cult). Results: The percentages of calvings correctly classifi ed by CART, CHAID, QUEST, and GLM were as follows: 35.14, 18.92, 19.82, and 43.24% (easy), 68.70, 73.91, 81.74, and 41.74% (moderate), and 77.27, 85.45, 73.64, and 81.82% (diffi cult), respectively. The most important factors aff ecting calving diffi culty were bull's rank (based on the mean calving diffi culty score of its daughters), calving age, farm category (based on its mean milk yield) and calving season. Conclusion: All classifi cation models were satisfactory and could predict the class of calving diffi culty.