{"title":"Observations about the Principal Components Analysis and Data Clustering Techniques in the Study of Medical Data","authors":"Cristina G. Dasc\u00e2lu, Corina Dima Cozma, Elena Carmen Cotrutz","volume":17,"journal":"International Journal of Medical and Health Sciences","pagesStart":162,"pagesEnd":167,"ISSN":"1307-6892","URL":"https:\/\/publications.waset.org\/pdf\/15595","abstract":"The medical data statistical analysis often requires the\nusing of some special techniques, because of the particularities of\nthese data. The principal components analysis and the data clustering\nare two statistical methods for data mining very useful in the medical\nfield, the first one as a method to decrease the number of studied\nparameters, and the second one as a method to analyze the\nconnections between diagnosis and the data about the patient-s\ncondition. In this paper we investigate the implications obtained from\na specific data analysis technique: the data clustering preceded by a\nselection of the most relevant parameters, made using the principal\ncomponents analysis. Our assumption was that, using the principal\ncomponents analysis before data clustering - in order to select and to\nclassify only the most relevant parameters \u2013 the accuracy of\nclustering is improved, but the practical results showed the opposite\nfact: the clustering accuracy decreases, with a percentage\napproximately equal with the percentage of information loss reported\nby the principal components analysis.","references":"[1] Chernick, M.R., Friis, R.H., Introductory Biostatistics for the Health\nSciences, John Wiley & Sons Publ., 2003.\n[2] Zhou, X.H., Obuchowski, N.A., McClish, D.K., Statistical Methods in\nDiagnostic Medicine, John Wiley & Sons Publ., 2002.\n[3] Saporta, G., \u253c\u00d7tef\u00e2nescu, M.V., Analiza datelor \u253c\u0192i informatic\u00e2, Ed.\nEconomic\u00e2, 1996 (in romanian).\n[4] C. Dasc\u00e2lu, Boiculese, L., \"The Usefulness of Algorithms Based on\nClustering in the Diagnosis Finding in Medical Practice\", in Lecture\nNotes of the ICB Seminars - Statistics and Clinical Practice, editors: L.\nBobrowski, J. Doroszewski, E. Marubini, N. Victor, Warsaw, 2000, pg.\n53 - 56.\n[5] Alsabti, K., Ranka, S., Singh, V., \"An Efficient K-Means Clustering\nAlgorithm\", in Proceedings of the 1st Workshop on High-Performance\nData Mining, 1998.\n[6] Dumitrescu, D., Teoria clasific\u00e2rii, Babe\u253c\u0192 - Bolyai University, Cluj -\nNapoca, 1991 (in romanian).","publisher":"World Academy of Science, Engineering and Technology","index":"Open Science Index 17, 2008"}