Multidimensional scaling by optimizing goodness of fit to a nonmetric hypothesis

摘要: Multidimensional scaling is the problem of representingn objects geometrically byn points, so that interpoint distances correspond in some sense to experimental dissimilarities between objects. In just what and should has been left rather vague most approaches, thus leaving these approaches logically incomplete. Our fundamental hypothesis are monotonically related. We define a quantitative, intuitively satisfying measure goodness fit this hypothesis. technique multidimensional compute configuration points which optimizes fit. A practical computer program for doing calculations described companion paper.