Unsupervised versus Supervised Identification of Prognostic Factors in Patients with Localized Retroperitoneal Sarcoma: A Data Clustering and Mahalanobis Distance Approach

Table 6

Summary of results. In the casewise method, 4 and 5 cases were excluded because of too many missing values in the supervised and unsupervised methods, respectively. In the mean substitution, no cases were excluded since missing values were substituted by their respective means. Accuracy was computed considering the number of misclassified cases on the total of analyzed cases (68, 67, or 72 in supervised casewise, unsupervised casewise, or mean substitution, resp.).

	Supervised		Unsupervised
	Casewise	Mean substitution	Casewise	Mean substitution
Incorrect classification	10	11	9	12
Missing values	4	/	5	/
% accuracy	85.3	84.7	86.6	83.3