Aplicación de redes complejas al estudio de datos de gestión sanitaria: una perspectiva desde la minería de datos

ABSTRACT: Healthcare management is one of the first-class issues in any modern society. At the same time, new data mining techniques and tools are hatching as a response to an ever growing number of so-called big data applications; in particular, complex networks and network science are consolidatin...

Descripción completa

Detalles Bibliográficos
Autor: Dehesa Cueto-Felgueroso, Javier de la
Tipo de recurso: tesis de maestría
Fecha de publicación:2014
País:España
Institución:Universidad de Cantabria (UC)
Repositorio:UCrea Repositorio Abierto de la Universidad de Cantabria
Idioma:inglés
OAI Identifier:oai:repositorio.unican.es:10902/5946
Acceso en línea:http://hdl.handle.net/10902/5946
Access Level:acceso abierto
Palabra clave:Healthcare
Big data
Complex networks
Bayesian networks
Sanidad
Redes complejas
Redes bayesianas
Descripción
Sumario:ABSTRACT: Healthcare management is one of the first-class issues in any modern society. At the same time, new data mining techniques and tools are hatching as a response to an ever growing number of so-called big data applications; in particular, complex networks and network science are consolidating as the tool of choice for the analysis of many non-relational unstructured data sources. In this work we put in contact these two worlds by analysing the characteristics of a diagnoses network emerged from national public healthcare system data. Different methodologies to extract and transform the data are explored, and several results evidencing interesting patterns in the data are presented. Additionally, a complementary analysis from the point of view of Bayesian networks is as well proposed, establishing a comparative between both approaches