Multidimensional scaling

Summary

Multidimensional scaling places objects as points in a low-dimensional space so that their distances match given dissimilarities as well as possible. Its input is only a table of pairwise dissimilarities, not coordinates. Classical multidimensional scaling does this exactly when the dissimilarities are Euclidean distances, using the eigenvectors of a centred matrix. It is used to visualise similarity data in two or three dimensions.