Livestock Application: Naïve Bayes for Diseases Forecast in a Bovine Production Application

Abstract

The Brazilian agriculture expansion, the increase in exportation and the search to guarantee food safety is increasingly the investment urgency on technologies focused on this activity. Data Mining and Machine Learning could be a vehicle to facilitate the fight against the spread of infective severity. In this paper, we applied the Naïve Bayes algorithm in a web mobile application (App) developed in the Outsystems platform using low code, to predict diseases in beef cattle production. It can store information to examine the health and ensure well-being of livestock. The project was conducted using Cross Industry Standard Process for Data mining (CRIPS-DM) methodology, composed by the phases: business understanding, data understanding, data preparation, modeling, and evaluation. The paper presents a case study in an extensive beef cattle farm in Brazil, that nowadays, is facing a common problem in its own agriculture, where there are increasing cases of bovine mortality due to several diseases. As a result, qualitative research to assessing the App usability was done with seven producers. But two of them didn’t answer because didn’t have computer access. They said the application is relevant, versatile, easy, pleasant to use and to understand, also stating that use of artificial intelligence is interesting.