Order Invariant Methods for Data Analysis
Robin Sibson · Journal of the Royal Statistical Society Series B (Statistical Methodology) · 1972
Summary Analysis of objects-by-attributes matrices is often carried out by first calculating a measure of pairwise similarity or dissimilarity between the objects. This paper investigates order invariance criteria which can be applied to methods for analysing such measures, and applies these criteria to methods for classification, scaling, and sequencing.
Cited by 142 works. doi:10.1111/j.2517-6161.1972.tb00910.x
Authors
References
- Construction of Phylogenetic Trees
- Multidimensional Scaling by Optimizing Goodness of Fit to a Nonmetric Hypothesis
- Theory and Methods of Scaling
- The Analysis of Proximities: Multidimensional Scaling with an Unknown Distance Function. I.
- Nonmetric Multidimensional Scaling: A Numerical Method
- The Application of Computers to Taxonomy
- A General Nonmetric Technique for Finding the Smallest Coordinate Space for a Configuration of Points
- SLINK: An optimally efficient algorithm for the single-link cluster method
- Mathematical Taxonomy
- Elementary Linkage Analysis for Isolating Orthogonal and Oblique Types and Typal Relevancies