Bibliographic citations
Cueva, J., Elguera, P., Vilela, G. (2019). Propuesta de un modelo predictivo para efectivizar el proceso de validación de la información de los sistemas de agua y saneamiento de los centros poblados del Perú [Trabajo de investigación, Universidad Peruana de Ciencias Aplicadas (UPC)]. http://hdl.handle.net/10757/625610
Cueva, J., Elguera, P., Vilela, G. Propuesta de un modelo predictivo para efectivizar el proceso de validación de la información de los sistemas de agua y saneamiento de los centros poblados del Perú [Trabajo de investigación]. PE: Universidad Peruana de Ciencias Aplicadas (UPC); 2019. http://hdl.handle.net/10757/625610
@mastersthesis{renati/1790424,
title = "Propuesta de un modelo predictivo para efectivizar el proceso de validación de la información de los sistemas de agua y saneamiento de los centros poblados del Perú",
author = "Vilela Girón, Guillermo Daniel",
publisher = "Universidad Peruana de Ciencias Aplicadas (UPC)",
year = "2019"
}
The Water and Sanitation System in Peru is essential for the development of the rural population. By 2030, the Peruvian government plans to cover 100% of drinking water and sanitation in the rural area. The Ministry of Housing, Construction and Sanitation – MCVS, through the National Rural Sanitation Program - PNSR promotes the application of rural water and sanitation policies, in coordination with the Regional Governments. The Water and Sanitation Diagnosis process of the MVCS collects the information from the population centers and then this information must be validated. In the validation of the information, the population centers are selected to be validated in a random way, because they can not really identify the population centers that enter inconsistent information. This negatively affects decision-making for the prioritization of construction projects and improvement of water and sanitation systems. In our research, with data mining techniques we have identified the profiles of the population centers that should be considered for the validation of their information, that is, the population centers that have inconsistent information and for that reason should be validated and corrected. To prove the aforementioned, we have considered the 9,937 populated centers of the Cusco region, applying the CRISP-DM methodology and the naive bayes algorithms, close neighbors, decision tree (Classification) and simple k-means (Clustering) that have given us allowed to obtain the models (predictive and descriptive) using the WEKA software.
This item is licensed under a Creative Commons License