Authors:
Manuel Santos
1
;
João Pereira
2
and
Álvaro Silva
3
Affiliations:
1
Universidade do Minho, Portugal
;
2
Escola Superior de Educação de Viana do Castelo, Portugal
;
3
Serviço de Cuidados Intensivos, Hospital Geral de Santo António, Portugal
Keyword(s):
Clinical Data Mining, Clustering, Knowledge Discovery from Databases, Artificial Neural Networks, Organ Failure, Mortality Predicting Models, Intermediate Outcomes, Intensive Medicine.
Related
Ontology
Subjects/Areas/Topics:
Artificial Intelligence
;
Biomedical Engineering
;
Business Analytics
;
Data Engineering
;
Data Mining
;
Databases and Information Systems Integration
;
Datamining
;
Enterprise Information Systems
;
Health Information Systems
;
Sensor Networks
;
Signal Processing
;
Soft Computing
Abstract:
Clustering is a technique widely applied in Data Mining problems due to the granularity, accuracy and adjustment of the models induced. Although the referred results, this approach generates a considerable large set of models, which difficult the comprehension, the visualization and the application to new cases. This paper presents a framework to deal with the enounced problem supported by a three-dimensional matrix structure. The usability and benefits of this instrument are demonstrated trough a case study in the area of intensive medicine.