Stochastic proximity embedding

A recently introduced algorithm, called stochastic proximity embedding (SPE), is a novel self-organizing scheme that addresses the key limitations of Isomap (isometric feature mapping) and LLE. SPE builds on the same geodesic principle first proposed and exploited in Isomap, but introduces two algorithmic advances SPE circumvents the calculation of estimated geodesic distances, and uses a pairwise refinement scheme that does not require the complete distance, or proximity, r, matrix. Due to these advances, the method scales linearly with the number of points. 

In contrast to MDS, SPE preserves exact distances between neighboring points and lower bounds between remote points, thns allowing the manifold to nnfold and reveal its true intrinsic dimensionality. Essentially, the method views the input distances between remote points as lower bonnds of their true geodesic distances, and uses them as a means to impose global stmctnre. 

---

Agrafiotis DK. Stochastic proximity embedding. J Comput Chem 2003 24 1215-1221. 

Agrafiolis DK (2003) Stochastic proximity embedding. J Comput Chem 24 1215-1221 Xue L, Stahura PL, Bajorath J (2004) Cell-based partitioning. In Chemoinformatics concepts, methods, and tools for drug discovery. Chapter 9. Humana, Totowa Wickens TD (2009) Multiway contingency tables analysis for the social sciences. Psychology, New York... 

---