SEGMENTATION OF BOVINE LACTATION CURVES, DURING THE FIRST THIRD OF LACTATION VIA CLUSTER AND LINEAL DISCRIMINANT ANALYSIS

Abstract

This paper presents the results of a segmentation study with, Hierarchical Cluster Analysis Ward's algorithm, applied to the data from 465 lactations of dairy cows in the Araucanía region. Only the first third of the lactation was considered whith data from measurement intervals of 30 days this, obtaining an analysis matrix of 465 rows (lactations) x 5 columns (checks every 30 days). Validation and evaluation of the cluster solution, based on the clustering of cases (individuals), was performed by linear discriminant analysis. A segment was obtained that allowed the dairy cows to be grouped obtainedinto 4 productive groups. The groups differed in production volumen, but mainly in the type of production in terms of the patterns of production, whith normal and atypical production curves. The implementation of the multivariate statistical tools, to analise of groups and patterns of bovine lactation curves, is very relevant and highly efficient, validated in this study by linear discriminant analysis.