Paper:
Language: English:

Comparison of Linguistic and Regular Hard C-Means in Postoperative Patient Data

Sansanee Auephanwiriyakul^*, and Nipon Theera-Umpon^**

*Computer Engineering Department, Chiang Mai University, Chiang Mai 50200, Thailand
**Electrical Engineering Department, Chiang Mai University, Chiang Mai 50200, Thailand

Received: July 30, 2004

Accepted: September 12, 2004

Keywords: linguistic hard c-means, hard c-means, postoperative

Journal ref: Journal of Advanced Computational Intelligence and Intelligent Informatics, Vol.8, No.6 pp. 599-605, 2004

Abstract

Classification is one of the main topics in pattern recognition. Generally, classification problems deal with numeric data. However, there are uncertainties in the data frequently, such as data collected in management questionnaires, postoperative patient data, etc. Hence, classification problems on linguistic data are challenging. In this paper, we compare the performance of the linguistic Hard C-Means (LHCM) with the regular Hard C-Means (HCM) through the postoperative patient data. We found that the LHCM performs better than the regular HCM in terms of the classification rate. In addition, we are able to interpret each cluster attribute by the input linguistic label.

[Notice]
* "Preview" is the first 2 pages of the article. You don't need the registration.
* To read the PDF file you will then need to download and install the Adobe Reader.
Adobe Reader is free and available for download here: