Data for: Automatic prediction of stress in piglets (Sus Scrofa) using infrared skin temperature

Published: 09-12-2019| Version 1 | DOI: 10.17632/2yxs3pk9fm.1
Irenilza de Alencar Naas,
Henry Ungaro,
Felipe Fonseca,
Alexandra Ferreira da Cordeiro,
Jair Abe,
Fabio Amaral


When working with bigdata, it is common to come across imprecise, conflicting, and lack of data. Extracting useful information from inaccurate data is a challenge for IT in general. Depending on the data source, conflicting data can be presented and the usual procedures do not allow the direct treatment of such data. The prompt is that contradictory information is as vital as other information or may even be the key to indicating crucial information. Although there are methods for handling fuziness, such methods are not able to adequately address the concept of inconsistency. In this essay, we present a method that can directly deal with the concepts of imprecision, inconsistency, and paracompleteness (lack of data) in a non-trivial way. Such methodology is based on the paraconsistent annotated logic E that allows manipulating such concepts. Based on this logic was built the para-analyzer algorithm, a logic analyzer that allows analyzing the concepts mentioned. Based on the para-analyzer, a small expert system was built, which then decides imprecise, contradictory and incomplete data. The suggested method can be used in a variety of subjects, especially where there is imprecision in the data, conflicting data, and lack of data.