Different concepts of relevance in a numerical taxonomic study

Abstract

THE term relevance was introduced by Cain and Harrison1 and it was modified by Sokal and Sneath3. During a taximetric study of the genus Salix, willows, I2 found it necessary to recognize three concepts of relevance, which I defined as follows: "OTU by OTU relevance" is the ratio of the number of characters used in the comparison of a given pair of OTUs (operational taxonomic units3) to the total number of characters used in the study. Characters for which information is missing in at least one of the OTUs are not used in the comparison. This is relevant in the sense of Sokal and Sneath3. "OTU relevance" is the ratio of the number of characters actually recorded in a particular OTU to the total number of characters used in the study. It is a measure of the amount of missing data in an OTU, whatever the cause. "Character relevance" is the ratio of the number of OTUs in which a real value for the character in question has been recorded to the total number of OTUs in the study. © 1968 Nature Publishing Group.