ELM-SOM: A Continuous Self-Organizing Map for Visualization

Renjie Hu; Venous Roshdibenam; Hans J Johnson; Emil Eirola; Anton Akusok; Yoan Miche; Kaj-Mikael Bjork; Amaury Lendasse

doi:10.1109/IJCNN.2018.8489268

Conference proceeding

ELM-SOM: A Continuous Self-Organizing Map for Visualization

Renjie Hu, Venous Roshdibenam, Hans J Johnson, Emil Eirola, Anton Akusok, Yoan Miche, Kaj-Mikael Bjork and Amaury Lendasse

2018 International Joint Conference on Neural Networks (IJCNN), Vol.2018-, pp.1-8

International Joint Conference on Neural Networks (IJCNN), 2018 (2018)

07/2018

DOI: 10.1109/IJCNN.2018.8489268

View Online

Abstract

This paper presents a novel dimensionality reduction technique: ELM-SOM. This technique preserves the intrinsic quality of Self-Organizing Maps (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 six diverse datasets. Regarding reconstruction error, ELM-SOM is comparable to SOM while bringing continuity.

Machine Learning

Neurons

Topology

Dimensionality reduction

Dimensionality Reduction Techniques

Extreme Learning Machines

Manifolds

Neural Networks

Self-organizing feature maps

Self-Organizing Maps

Training

Urban areas

Visualization

Details

Title: Subtitle: ELM-SOM: A Continuous Self-Organizing Map for Visualization
Creators: Renjie Hu - University of Iowa
Venous Roshdibenam - University of Iowa
Hans J Johnson - University of Iowa
Emil Eirola - Arcada University of Applied Sciences
Anton Akusok - Arcada University of Applied Sciences
Yoan Miche - Bell Labs
Kaj-Mikael Bjork - Arcada University of Applied Sciences
Amaury Lendasse - University of Iowa
Resource Type: Conference proceeding
Publication Details: 2018 International Joint Conference on Neural Networks (IJCNN), Vol.2018-, pp.1-8
Conference: International Joint Conference on Neural Networks (IJCNN), 2018 (2018)
DOI: 10.1109/IJCNN.2018.8489268
eISSN: 2161-4407
Publisher: IEEE
Language: English
Date published: 07/2018
Academic Unit: Roy J. Carver Department of Biomedical Engineering; Electrical and Computer Engineering; Psychiatry; Industrial and Systems Engineering
Record Identifier: 9984185467002771

Metrics

28 Record Views