Flexible self-organizing maps in kohonen 3.0

Research output: Contribution to journalArticleAcademicpeer-review

9 Citations (Scopus)

Abstract

Self-organizing maps (SOMs) are popular tools for grouping and visualizing data in many areas of science. This paper describes recent changes in package kohonen, implementing several different forms of SOMs. These changes are primarily focused on making the package more useable for large data sets. Memory consumption has decreased dramatically, amongst others, by replacing the old interface to the underlying compiled code by a new one relying on Rcpp. The batch SOM algorithm for training has been added in both sequential and parallel forms. A final important extension of the package’s repertoire is the possibility to define and use data-dependent distance functions, extremely useful in cases where standard distances like the Euclidean distance are not appropriate. Several examples of possible applications are presented.

Original languageEnglish
Number of pages18
JournalJournal of Statistical Software
Volume87
Issue number7
DOIs
Publication statusPublished - 12 Nov 2018

Fingerprint

Self organizing maps
Self-organizing Map
Dependent Data
Distance Function
Euclidean Distance
Large Data Sets
Grouping
Batch
Data storage equipment
Self-organizing map
Form

Keywords

  • Distance functions
  • Parallellization, R.
  • Self-organizing maps

Cite this

@article{1a112f9de2a74c08be76a1f1ae16007a,
title = "Flexible self-organizing maps in kohonen 3.0",
abstract = "Self-organizing maps (SOMs) are popular tools for grouping and visualizing data in many areas of science. This paper describes recent changes in package kohonen, implementing several different forms of SOMs. These changes are primarily focused on making the package more useable for large data sets. Memory consumption has decreased dramatically, amongst others, by replacing the old interface to the underlying compiled code by a new one relying on Rcpp. The batch SOM algorithm for training has been added in both sequential and parallel forms. A final important extension of the package’s repertoire is the possibility to define and use data-dependent distance functions, extremely useful in cases where standard distances like the Euclidean distance are not appropriate. Several examples of possible applications are presented.",
keywords = "Distance functions, Parallellization, R., Self-organizing maps",
author = "Ron Wehrens and Johannes Kruisselbrink",
year = "2018",
month = "11",
day = "12",
doi = "10.18637/jss.v087.i07",
language = "English",
volume = "87",
journal = "The journal of statistical software",
issn = "1548-7660",
publisher = "University of California at Los Angeles",
number = "7",

}

Flexible self-organizing maps in kohonen 3.0. / Wehrens, Ron; Kruisselbrink, Johannes.

In: Journal of Statistical Software, Vol. 87, No. 7, 12.11.2018.

Research output: Contribution to journalArticleAcademicpeer-review

TY - JOUR

T1 - Flexible self-organizing maps in kohonen 3.0

AU - Wehrens, Ron

AU - Kruisselbrink, Johannes

PY - 2018/11/12

Y1 - 2018/11/12

N2 - Self-organizing maps (SOMs) are popular tools for grouping and visualizing data in many areas of science. This paper describes recent changes in package kohonen, implementing several different forms of SOMs. These changes are primarily focused on making the package more useable for large data sets. Memory consumption has decreased dramatically, amongst others, by replacing the old interface to the underlying compiled code by a new one relying on Rcpp. The batch SOM algorithm for training has been added in both sequential and parallel forms. A final important extension of the package’s repertoire is the possibility to define and use data-dependent distance functions, extremely useful in cases where standard distances like the Euclidean distance are not appropriate. Several examples of possible applications are presented.

AB - Self-organizing maps (SOMs) are popular tools for grouping and visualizing data in many areas of science. This paper describes recent changes in package kohonen, implementing several different forms of SOMs. These changes are primarily focused on making the package more useable for large data sets. Memory consumption has decreased dramatically, amongst others, by replacing the old interface to the underlying compiled code by a new one relying on Rcpp. The batch SOM algorithm for training has been added in both sequential and parallel forms. A final important extension of the package’s repertoire is the possibility to define and use data-dependent distance functions, extremely useful in cases where standard distances like the Euclidean distance are not appropriate. Several examples of possible applications are presented.

KW - Distance functions

KW - Parallellization, R.

KW - Self-organizing maps

U2 - 10.18637/jss.v087.i07

DO - 10.18637/jss.v087.i07

M3 - Article

VL - 87

JO - The journal of statistical software

JF - The journal of statistical software

SN - 1548-7660

IS - 7

ER -