Analysis of sea water infiltration in a sewage treatment plant using Random Forests and variable importance measures
Metadatos
Mostrar el registro completo del ítemcomunitat-uji-handle:10234/158176
comunitat-uji-handle2:10234/71345
comunitat-uji-handle3:10234/97762
comunitat-uji-handle4:
TFG-TFMMetadatos
Título
Analysis of sea water infiltration in a sewage treatment plant using Random Forests and variable importance measuresAutoría
Tutor/Supervisor; Universidad.Departamento
Epifanio López, Irene; Universitat Jaume I. Departament de MatemàtiquesFecha de publicación
2017-11-23Editor
Universitat Jaume IResumen
In ltration of salt water in water-treatment plants is a current challenge in Spanish coasts.
This is a problem because when the sea water enters into the plant, it damages the lters and
compromise the quality of ... [+]
In ltration of salt water in water-treatment plants is a current challenge in Spanish coasts.
This is a problem because when the sea water enters into the plant, it damages the lters and
compromise the quality of the water ltration. In order to detect this water in ltration, the
usual approach is to measure conductivity of the
ows (ability to conduct electricity).
The work subsequently described in this document is a project funded by a research internship
of the Universitat Jaume I's C atedra FACSA de Innovaci on del Ciclo Integral del Agua. The
goal of this project is to detect when a saltwater in ltration has occurred and to detect the most
relevant variables which are related with the rising of water conductivity.
The approach chosen to deal with this problem has been the technique of Random Forests, a
family of algorithm based on decision trees. The reasons of this elections are mainly the
exibility
with respect missing data that Random Forests allow; the \black box"-like behaviour that does
not need an a priori knowledge of the data structure; and the ability to explore the variables'
importance through several measures.
In this project the mentioned methodology will be explained in detail, as well as the results
obtained and the conclusions that follows them. [-]
Palabras clave / Materias
Descripción
Treball de Fi de Màster Universitari en Matemàtica Computacional (Pla de 2013). Codi: SIQ527. Curs 2016/2017
Tipo de documento
info:eu-repo/semantics/masterThesisDerechos de acceso
info:eu-repo/semantics/openAccess
Aparece en las colecciones
El ítem tiene asociados los siguientes ficheros de licencia: