Descoberta de Conhecimento com Mineração de Dados Aplicada à Ovinocultura

Abstract

Nowadays, companies tend to maintain great and growing volumes of data. Eventually, these datasets may reach massive amounts and this scenario is convenient for the discovery of knowledge in databases. In this context, data mining is applied through its functionalities and algorithmic implementations that are specific for that purpose. This research is about the results of knowledge discovery tests in sheep breeding with the application of data mining, utilizing pre-processing of the provided dataset in each specific experiment. Association, classification and clustering tasks were performed with the 1R, C4.5, CART and k-means algorithms with the Waikato Environment for Knowledge Analysis (WEKA) tool and enabled the discovery of out-of-pattern activity evidence, as well as information that is potentially relevant for the sheep breeding business domain.