Comparing decision tree and optimal risk pattern mining for analysing emergency ultra short stay unit data

Date

2008

Authors

Petrus, K.
Li, J.
Fahey, P.

Editors

Advisors

Journal Title

Journal ISSN

Volume Title

Type:

Conference paper

Citation

Proceedings of the seventh international conference on machine learning and cybernetics, 2008, vol.1, pp.234-239

Statement of Responsibility

Conference Name

7th International Conference on Machine Learning and Cybernetics (12 Jul 2008 - 15 Jul 2008 : Kunming, China)

Abstract

A data set contains patient records of Ultra Short Stay Unit (USSU) at emergency department at Toowoomba Base Hospital. Some patients were admitted to the hospital for further treatment after a long stay at USSU and other patients were discharged after a short stay at USSU. In most hospitals the USSU is not enough for large demand, and there will be better utilisation of the unit if medical professionals know what types of patients are more likely to be hospitalised before any treatment at USSU. Two data mining methods; decision trees and optimal risk pattern mining, have been applied on the data to explore cohorts of patients who are more likely to be admitted to the hospital. Results show that decision tree method is inadequate for finding understandable patterns, and that optimal risk pattern mining method is good for mining meaningful patterns for medical practitioners. © 2008 IEEE.

School/Discipline

Dissertation Note

Provenance

Description

Access Status

Rights

Copyright status unknown

License

Grant ID

Call number

Persistent link to this record