ELM-SOM+: A continuous mapping for visualization

Renjie Hu; Karl Ratner; Edward Ratner; Yoan Miche; Kaj Mikael Björk; Amaury Lendasse

doi:10.1016/j.neucom.2019.06.093

ELM-SOM+: A continuous mapping for visualization

Renjie Hu^*, Karl Ratner, Edward Ratner, Yoan Miche, Kaj Mikael Björk, Amaury Lendasse

^*Corresponding author for this work

Finance, Helsinki

Research output: Contribution to journal › Article › Scientific › peer-review

10 Citations (Scopus)

Abstract

This paper presents a novel dimensionality reduction technique based on ELM and SOM: ELM-SOM+. This technique preserves the intrinsic quality of Self-Organizing Map (SOM): it is nonlinear and suitable for big data. It also brings continuity to the projection using two Extreme Learning Machine (ELM) models, the first one to perform the dimensionality reduction and the second one to perform the reconstruction. ELM-SOM+ is tested successfully on nine diverse datasets. Regarding reconstruction error, the new methodology shows considerable improvement over SOM and brings continuity.

Original language	English
Peer-reviewed scientific journal	Neurocomputing
Volume	365
Pages (from-to)	147-156
Number of pages	10
ISSN	0925-2312
DOIs	https://doi.org/10.1016/j.neucom.2019.06.093
Publication status	Published - 06.11.2019
MoE publication type	A1 Journal article - refereed

Keywords

Self-Organizing Maps
Visualization
512 Business and Management
Dimensionality reduction techniques
Visualization
Extreme Learning Machines
Self-Organizing Maps
Machine learning
Neural networks

Access to Document

10.1016/j.neucom.2019.06.093

OpenUrl availability

Full text

Cite this

@article{1e45958f450445a28bd4244c9ff17b43,

title = "ELM-SOM+: A continuous mapping for visualization",

abstract = "This paper presents a novel dimensionality reduction technique based on ELM and SOM: ELM-SOM+. This technique preserves the intrinsic quality of Self-Organizing Map (SOM): it is nonlinear and suitable for big data. It also brings continuity to the projection using two Extreme Learning Machine (ELM) models, the first one to perform the dimensionality reduction and the second one to perform the reconstruction. ELM-SOM+ is tested successfully on nine diverse datasets. Regarding reconstruction error, the new methodology shows considerable improvement over SOM and brings continuity.",

keywords = "Self-Organizing Maps, Visualization, 512 Business and Management, Dimensionality reduction techniques, Visualization, Extreme Learning Machines, Self-Organizing Maps, Machine learning, Neural networks",

author = "Renjie Hu and Karl Ratner and Edward Ratner and Yoan Miche and Bj{\"o}rk, {Kaj Mikael} and Amaury Lendasse",

year = "2019",

month = nov,

day = "6",

doi = "10.1016/j.neucom.2019.06.093",

language = "English",

volume = "365",

pages = "147--156",

journal = "Neurocomputing",

issn = "0925-2312",

publisher = "Elsevier",

}

TY - JOUR

T1 - ELM-SOM+

T2 - A continuous mapping for visualization

AU - Hu, Renjie

AU - Ratner, Karl

AU - Ratner, Edward

AU - Miche, Yoan

AU - Björk, Kaj Mikael

AU - Lendasse, Amaury

PY - 2019/11/6

Y1 - 2019/11/6

N2 - This paper presents a novel dimensionality reduction technique based on ELM and SOM: ELM-SOM+. This technique preserves the intrinsic quality of Self-Organizing Map (SOM): it is nonlinear and suitable for big data. It also brings continuity to the projection using two Extreme Learning Machine (ELM) models, the first one to perform the dimensionality reduction and the second one to perform the reconstruction. ELM-SOM+ is tested successfully on nine diverse datasets. Regarding reconstruction error, the new methodology shows considerable improvement over SOM and brings continuity.

AB - This paper presents a novel dimensionality reduction technique based on ELM and SOM: ELM-SOM+. This technique preserves the intrinsic quality of Self-Organizing Map (SOM): it is nonlinear and suitable for big data. It also brings continuity to the projection using two Extreme Learning Machine (ELM) models, the first one to perform the dimensionality reduction and the second one to perform the reconstruction. ELM-SOM+ is tested successfully on nine diverse datasets. Regarding reconstruction error, the new methodology shows considerable improvement over SOM and brings continuity.

KW - Self-Organizing Maps

KW - Visualization

KW - 512 Business and Management

KW - Dimensionality reduction techniques

KW - Visualization

KW - Extreme Learning Machines

KW - Self-Organizing Maps

KW - Machine learning

KW - Neural networks

UR - http://www.scopus.com/inward/record.url?scp=85069956302&partnerID=8YFLogxK

U2 - 10.1016/j.neucom.2019.06.093

DO - 10.1016/j.neucom.2019.06.093

M3 - Article

AN - SCOPUS:85069956302

SN - 0925-2312

VL - 365

SP - 147

EP - 156

JO - Neurocomputing

JF - Neurocomputing

ER -

ELM-SOM+: A continuous mapping for visualization

Abstract

Keywords

Access to Document

OpenUrl availability

Other files and links

Fingerprint

Cite this