• Corpus ID: 220714117

# Multidimensional Scaling for Big Data

@article{Delicado2020MultidimensionalSF,
title={Multidimensional Scaling for Big Data},
author={Pedro Delicado and Cristian Pachon-Garcia},
journal={arXiv: Computation},
year={2020}
}
• Published 23 July 2020
• Computer Science
• arXiv: Computation
We present a set of algorithms for \textit{Multidimensional Scaling} (MDS) to be used with large datasets. MDS is a statistic tool for reduction of dimensionality, using as input a distance matrix of dimensions $n \times n$. When $n$ is large, classical algorithms suffer from computational problems and MDS configuration can not be obtained. In this paper we address these problems by means of three algorithms: Divide and Conquer MDS, Fast MDS and MDS based on Gower interpolation (the first and…
1 Citations

## Figures and Tables from this paper

Multidimensional scaling for Big Data
This study points out that Fast MDS and MDS based on Gower interpolation are appropriated to use when n is large and Divide and Conquer MDS is the best method that captures the variance of the original data.

## References

SHOWING 1-10 OF 14 REFERENCES
Modern Multidimensional Scaling: Theory and Applications
• Computer Science
• 1997
The four Purposes of Multidimensional Scaling, Special Solutions, Degeneracies, and Local Minima, and Avoiding Trivial Solutions in Unfolding are explained.
R: A language and environment for statistical computing.
Copyright (©) 1999–2012 R Foundation for Statistical Computing. Permission is granted to make and distribute verbatim copies of this manual provided the copyright notice and this permission notice
Applied Multivariate Statistical Analysis (5th ed.)
• 2002
Principles of Multivariate Analysis (Revised ed.), Volume 23 of Oxford Statistical Science Series
• 2000
Numerical Linear Algebra
• Computer Science
Encyclopedia of Parallel Computing
• 2011
EMNIST: Extending MNIST to handwritten letters
• Computer Science
2017 International Joint Conference on Neural Networks (IJCNN)
• 2017
A variant of the full NIST dataset is introduced, which is called Extended MNIST (EMNIST), which follows the same conversion paradigm used to create the MNIST dataset, and one that shares the same image structure and parameters as the original MNIST task, allowing for direct compatibility with all existing classifiers and systems.
A general method for solving divide-and-conquer recurrences
• Computer Science
SIGA
• 1980
A unifying method for solving recurrence relations of the form T(n) = kT(n/c) + f( n) is described that is both general in applicability and easy to apply.