Diagnosing patients with a combination of principal component analysis and case based reasoning

Authors:
Carles Pous;Dani Caballero;Beatriz Lopez
Affiliations:
(Correspd. Tel.: +34 972418390/ Fax: +34 972418976/ E-mail: carles.pous@udg.edu) University of Girona, Campus de Montilivi, Girona, Spain;University of Girona, Campus de Montilivi, Girona, Spain;University of Girona, Campus de Montilivi, Girona, Spain
Venue:
International Journal of Hybrid Intelligent Systems - Data Mining and Hybrid Intelligent Systems
Year:
2009

Citing 5
Cited 1

Case‐Based Reasoning: an overview

AI Communications
Case-Based Reasoning Combined with Information Entropy and Principal Component Analysis for Short-Term Load Forecasting

CIS '07 Proceedings of the 2007 International Conference on Computational Intelligence and Security
Case-based reasoning in the health sciences: What's next?

Artificial Intelligence in Medicine
Maximum likelihood hebbian learning based Retrieval method for CBR systems

ICCBR'03 Proceedings of the 5th international conference on Case-based reasoning: Research and Development
Covariance and PCA for categorical variables

PAKDD'05 Proceedings of the 9th Pacific-Asia conference on Advances in Knowledge Discovery and Data Mining

Intelligence techniques for prostate ultrasound image analysis

International Journal of Hybrid Intelligent Systems

Quantified Score

Hi-index	0.00

Visualization

Abstract

This paper addresses the application of a principal component analysis (PCA) of categorical data prior to diagnosing a patients dataset using a case-based reasoning (CBR) system. The particularity is that the standard PCA techniques are designed to deal with numerical attributes, but our medical dataset contains many categorical data and alternative methods such as RS-PCA are required. Thus, we propose to hybridize RS-PCA (regular simplex PCA) and a simple CBR system. Results show how the hybrid system, when diagnosing a medical dataset, produces results similar to the ones obtained when using the original attributes. These results are quite promising since they allow diagnosis with less computation effort and memory storage.