Descoberta de padrões no período reprodutivo de macacos da espécie Saimiri Collinsi através da mineração de dados

Abstract

Even today, some of the basic analyzes have been performed manually or almost without using specialized technologies, such as the Veterinary Medicine researchers of the Federal University of Pará - Campus Castanhal, who carry out their analyzes through statistical calculations using spreadsheets, taking enough time to analyze them until useful information is actually detected. Thus, the present work describes application knowledge discovery in Database’s process under dataset of researchers of Veterinary Medicine to solve a real problem related to the detection of patterns through the physical and morphological characteristics of monkeys of the species Saimiri Collinsi living in captivity. With emphasis on the data mining stage, the following assumptions are determined: determine which physical characteristics (arm and chest skinfolds, weight and testicular volume) have a greater influence on the reproductive season of animals. After, a new analysis under the parameters of seminal quality is made as the objective is to identify the best period for the collection of semen of the monkeys. All this process was performed through the association rules through the Apriori and the FP-Growth algorithms, for a better performance of the results obtained through an algorithmic comparison. It is worth mentioning that there is a need to generate synthetic data (DS) due to the small amount of data available to perform the algorithm training. Thus, each result obtained in this research was analyzed and evaluated together with the experts of the area, so that there were no doubts or misinterpretations thus ensuring their integrity.