Conceptual Clustering and Analysis of Data from Gynecological Database

Martinovska, Cveta (2010) Conceptual Clustering and Analysis of Data from Gynecological Database. In: ICT Innovations 2009. Springer-Verlag, Berlin Heidelberg, pp. 45-54. ISBN 978-3-642-10780-1

Full text not available from this repository. (Request a copy)
Official URL: http://www.springerlink.com/content/M4343573L2KV68...

Abstract

The aim of this work is to propose a methodology for classifying, analyzing and visualizing data of patients with different symptoms from gynecological database. The application implements a variant of WITT algorithm for conceptual clustering. Pre-clustering algorithm is proposed that includes a tradeoff between overlapping of the initial clusters and displacing the center of clusters far away from the region of great density. To overcome the problem with weak correlation different coding schemes for cases are tested. Successful approach was to take square root of attribute value intervals to achieve the intervals with different sizes. Two different datasets from gynecological database are used: data related to polycystic ovary syndrome and data relevant to diagnose pre-eclampsia.

Item Type: Book Section
Subjects: Natural sciences > Computer and information sciences
Divisions: Faculty of Computer Science
Depositing User: Cveta Martinovska Bande
Date Deposited: 06 Nov 2012 15:11
Last Modified: 06 Nov 2012 15:11
URI: http://eprints.ugd.edu.mk/id/eprint/515

Actions (login required)

View Item View Item