Model of neural networks: probabilistic prediction of floods in banana agricultural field

• Autores: Trujillo Moreno, Holiver, Gómez Márquez, Renzon Javier, Cano Lengua, Miguel Ángel, Andrade Arenas, Laberiano
• Formato: artículo
• Fecha de Publicación: 2023
• Institución: Universidad Tecnológica del Perú
• Repositorio: UTP-Institucional
• Lenguaje: inglés
• OAI Identifier: oai:repositorio.utp.edu.pe:20.500.12867/6881
• Enlace del recurso: https://hdl.handle.net/20.500.12867/6881; http://doi.org/10.14445/22315381/IJETT-V71I1P211
• Nivel de acceso: acceso abierto
• Materia: Artificial neural networks; Predictive modelling; Machine learning; Flood risk; Agriculture; ttps://purl.org/pe-repo/ocde/ford#1.02.00; https://purl.org/pe-repo/ocde/ford#4.01.00

During the latest events caused by climate change and the current of the child, Peru has been affected by these natural disasters, such as the flood, which directly affect the Peruvian economy and especially the department of Piura. To prevent and mitigate the problems that affect the department of Piura with respect to flooding, the development of a probabilistic system has been proposed with the use of machine learning that will allow us to prevent possible climatic changes and avoid material damage to the area based on predictions. Likewise, the data found in the repository of the free data web page provided by SENAMHI will be extracted to be reused internally and can contribute to the development of the application through neural networks that will facilitate the use of the data. Given this, it has been decided to use the data scientific method, which consists of 10 phases that allow us to identify the main points that contribute to the model of the proposal. This allows us to carry out the necessary validations to make the proposed system feasible. To obtain, as a result, a model that can predict and give warning about the threat of flooding based on the weather behavior of the area. In addition, it is concluded that the prediction models with the help of artificial intelligence tools have better efficiency in terms of forecasts.