Dimensionality Reduction for Classification through Visualisation Using L1SNE

Lennon Cook; Junbin Gao

Dimensionality Reduction for Classification through Visualisation Using L1SNE

Lennon Cook
, Junbin Gao

Research output: Book chapter/Published conference paper › Conference paper › peer-review

Abstract

Dimensionality Reduction algorithms have wide precedent for use in preprocessing for classification problems. This paper presents a new algorithm, based on a modification to Stochastic Neighbour Embedding and t-Distributed SNE to use the Laplacian distribution instead of, respectively, the Gaussian Distribution and a mismatched pair of the Gaussian Distribution and Student's t-Distribution. Experimental results are presented to demonstrate that this modification yields improvement.

Original language	English
Title of host publication	Australian Joint Conference on Artificial Intelligence
Editors	Jiuyong Li
Place of Publication	Germany
Publisher	Springer
Pages	204-212
Number of pages	9
Publication status	Published - 2010
Event	AI 2010 23rd conference - Adelaide, SA, Australia Duration: 07 Dec 2010 → 10 Dec 2010

Conference

Conference	AI 2010 23rd conference
Country/Territory	Australia
Period	07/12/10 → 10/12/10

Cite this

@inproceedings{06faf0595beb488ea85601d5a70a0af8,

title = "Dimensionality Reduction for Classification through Visualisation Using L1SNE",

abstract = "Dimensionality Reduction algorithms have wide precedent for use in preprocessing for classification problems. This paper presents a new algorithm, based on a modification to Stochastic Neighbour Embedding and t-Distributed SNE to use the Laplacian distribution instead of, respectively, the Gaussian Distribution and a mismatched pair of the Gaussian Distribution and Student's t-Distribution. Experimental results are presented to demonstrate that this modification yields improvement.",

keywords = "Dimensionality Reduction",

author = "Lennon Cook and Junbin Gao",

note = "Imported on 03 May 2017 - DigiTool details were: publisher = Germany: Springer, 2010. editor/s (773b) = Jiuyong Li; Event dates (773o) = 7-10 Dec 2010; Parent title (773t) = Australian Joint Conference on Artificial Intelligence. ISSNs: 0302-9743; ; AI 2010 23rd conference ; Conference date: 07-12-2010 Through 10-12-2010",

year = "2010",

language = "English",

pages = "204--212",

editor = "Jiuyong Li",

booktitle = "Australian Joint Conference on Artificial Intelligence",

publisher = "Springer",

address = "United States",

}

TY - GEN

T1 - Dimensionality Reduction for Classification through Visualisation Using L1SNE

AU - Cook, Lennon

AU - Gao, Junbin

N1 - Imported on 03 May 2017 - DigiTool details were: publisher = Germany: Springer, 2010. editor/s (773b) = Jiuyong Li; Event dates (773o) = 7-10 Dec 2010; Parent title (773t) = Australian Joint Conference on Artificial Intelligence. ISSNs: 0302-9743;

PY - 2010

Y1 - 2010

N2 - Dimensionality Reduction algorithms have wide precedent for use in preprocessing for classification problems. This paper presents a new algorithm, based on a modification to Stochastic Neighbour Embedding and t-Distributed SNE to use the Laplacian distribution instead of, respectively, the Gaussian Distribution and a mismatched pair of the Gaussian Distribution and Student's t-Distribution. Experimental results are presented to demonstrate that this modification yields improvement.

AB - Dimensionality Reduction algorithms have wide precedent for use in preprocessing for classification problems. This paper presents a new algorithm, based on a modification to Stochastic Neighbour Embedding and t-Distributed SNE to use the Laplacian distribution instead of, respectively, the Gaussian Distribution and a mismatched pair of the Gaussian Distribution and Student's t-Distribution. Experimental results are presented to demonstrate that this modification yields improvement.

KW - Dimensionality Reduction

M3 - Conference paper

SP - 204

EP - 212

BT - Australian Joint Conference on Artificial Intelligence

A2 - Li, Jiuyong

PB - Springer

CY - Germany

T2 - AI 2010 23rd conference

Y2 - 7 December 2010 through 10 December 2010

ER -

Dimensionality Reduction for Classification through Visualisation Using L1SNE

Abstract

Conference

Fingerprint

Cite this