On a General Transformation Making a Dissimilarity Matrix Euclidean

When a dissimilarity matrix cannot be represented in a Euclidean space, it is possible to make it Euclidean by means of suitable transformations of the original dissimilarity values. In this paper we discuss some interesting properties of a class of transformations based on adding a specific squared Euclidean distance to the initial dissimilarity. An erratum to this article is available at .