Predictive model of water potability through a decision tree in Artificial Intelligence

• Autores: Zevallos Apaza, Angel Alexis, Onque Gárate, Sofía Sair, Canaza Cuadros, Arian Eduardo Javier, Choqueneira Ccasa, Paulina Miriam
• Formato: artículo
• Fecha de Publicación: 2022
• Institución: Universidad La Salle
• Repositorio: Revistas - Universidad La Salle
• Lenguaje: español
• OAI Identifier: oai:ojs.revistas.ulasalle.edu.pe:article/72
• Enlace del recurso: https://revistas.ulasalle.edu.pe/innosoft/article/view/72; https://doi.org/10.48168/innosoft.s9.a72; https://purl.org/42411/s9/a72; https://n2t.net/ark:/42411/s9/a72
• Nivel de acceso: acceso abierto
• Materia: Drinking water; artificial intelligence; decision tree; Agua potable; inteligencia artificial; árbol de decisión

The objective of this work was to use the decision tree technique to define a model capable of predicting water potability. To evaluate the performance of the decision tree classification, a dataset extracted from Kaggle was used, which has 3276 water samples divided by the potability variable. Applying the Pandas and Scikit Learn libraries, a model based on a decision tree evaluated with the metrics of precision, accuracy, completeness, and F1 score was defined, achieving 0.77, 0.80, 0.85, and 0.81, respectively.