Applying neural networks to determine the socio-environmental factors responsible for potable water consumption

This study aimed to identify the variables that most influence the potable water consumption of the State of Paraná, Brazil. The study attempted to model the consumption of water using Artificial Neural Networks (ANN) associated with the extraction of knowledge. The results indicate that the water consumption of the State of Paraná is directly related to socio-environmental factors. However, when compared to the State of Minas Gerais (MG - Brazil), it was found that the decision tree generated does not reflect the same variables, indicating specific behaviors of that state. The results and analysis presented in this work may be used by campaigns such as the National Program to Combat Water Wastage (PNCDA)-BR and the Millennium Project established by the United Nations to reduce the inappropriate use of water.