P-ISOMAP: An efficient parametric update for ISOMAP for visual analytics

Jaegul Choo, Chandan K. Reddy, Hanseung Lee, Haesun Park

Research output: Contribution to conferencePaper

3 Citations (Scopus)

Abstract

One of the most widely-used nonlinear data embedding methods is ISOMAP. Based on a manifold learning framework, ISOMAP has a parameter k or ε that controls how many edges a neighborhood graph has. However, a suitable parameter value is often difficult to determine because of a time-consuming optimization process based on certain criteria, which may not be clearly justified. When ISOMAP is used to visualize data, users might want to test different parameter values in order to gain various insights about data, but such interaction between humans and such visualizations requires reasonably efficient updating, even for large-scale data. To tackle these problems, we propose an efficient updating algorithm for ISOMAP with parameter changes, called p-ISOMAP. We present not only a complexity analysis but also an empirical running time comparison, which show the advantage of p-ISOMAP. We also show interesting visualization applications of p-ISOMAP and demonstrate how to discover various characteristics of data through visualizations using different parameter values.

Original languageEnglish
Pages502-513
Number of pages12
Publication statusPublished - 2010 Dec 1
Externally publishedYes
Event10th SIAM International Conference on Data Mining, SDM 2010 - Columbus, OH, United States
Duration: 2010 Apr 292010 May 1

Conference

Conference10th SIAM International Conference on Data Mining, SDM 2010
CountryUnited States
CityColumbus, OH
Period10/4/2910/5/1

Fingerprint

Visualization

ASJC Scopus subject areas

  • Software

Cite this

Choo, J., Reddy, C. K., Lee, H., & Park, H. (2010). P-ISOMAP: An efficient parametric update for ISOMAP for visual analytics. 502-513. Paper presented at 10th SIAM International Conference on Data Mining, SDM 2010, Columbus, OH, United States.

P-ISOMAP : An efficient parametric update for ISOMAP for visual analytics. / Choo, Jaegul; Reddy, Chandan K.; Lee, Hanseung; Park, Haesun.

2010. 502-513 Paper presented at 10th SIAM International Conference on Data Mining, SDM 2010, Columbus, OH, United States.

Research output: Contribution to conferencePaper

Choo, J, Reddy, CK, Lee, H & Park, H 2010, 'P-ISOMAP: An efficient parametric update for ISOMAP for visual analytics', Paper presented at 10th SIAM International Conference on Data Mining, SDM 2010, Columbus, OH, United States, 10/4/29 - 10/5/1 pp. 502-513.
Choo J, Reddy CK, Lee H, Park H. P-ISOMAP: An efficient parametric update for ISOMAP for visual analytics. 2010. Paper presented at 10th SIAM International Conference on Data Mining, SDM 2010, Columbus, OH, United States.
Choo, Jaegul ; Reddy, Chandan K. ; Lee, Hanseung ; Park, Haesun. / P-ISOMAP : An efficient parametric update for ISOMAP for visual analytics. Paper presented at 10th SIAM International Conference on Data Mining, SDM 2010, Columbus, OH, United States.12 p.
@conference{7fb2ef8aa1944ff2a5faced8696665c7,
title = "P-ISOMAP: An efficient parametric update for ISOMAP for visual analytics",
abstract = "One of the most widely-used nonlinear data embedding methods is ISOMAP. Based on a manifold learning framework, ISOMAP has a parameter k or ε that controls how many edges a neighborhood graph has. However, a suitable parameter value is often difficult to determine because of a time-consuming optimization process based on certain criteria, which may not be clearly justified. When ISOMAP is used to visualize data, users might want to test different parameter values in order to gain various insights about data, but such interaction between humans and such visualizations requires reasonably efficient updating, even for large-scale data. To tackle these problems, we propose an efficient updating algorithm for ISOMAP with parameter changes, called p-ISOMAP. We present not only a complexity analysis but also an empirical running time comparison, which show the advantage of p-ISOMAP. We also show interesting visualization applications of p-ISOMAP and demonstrate how to discover various characteristics of data through visualizations using different parameter values.",
author = "Jaegul Choo and Reddy, {Chandan K.} and Hanseung Lee and Haesun Park",
year = "2010",
month = "12",
day = "1",
language = "English",
pages = "502--513",
note = "10th SIAM International Conference on Data Mining, SDM 2010 ; Conference date: 29-04-2010 Through 01-05-2010",

}

TY - CONF

T1 - P-ISOMAP

T2 - An efficient parametric update for ISOMAP for visual analytics

AU - Choo, Jaegul

AU - Reddy, Chandan K.

AU - Lee, Hanseung

AU - Park, Haesun

PY - 2010/12/1

Y1 - 2010/12/1

N2 - One of the most widely-used nonlinear data embedding methods is ISOMAP. Based on a manifold learning framework, ISOMAP has a parameter k or ε that controls how many edges a neighborhood graph has. However, a suitable parameter value is often difficult to determine because of a time-consuming optimization process based on certain criteria, which may not be clearly justified. When ISOMAP is used to visualize data, users might want to test different parameter values in order to gain various insights about data, but such interaction between humans and such visualizations requires reasonably efficient updating, even for large-scale data. To tackle these problems, we propose an efficient updating algorithm for ISOMAP with parameter changes, called p-ISOMAP. We present not only a complexity analysis but also an empirical running time comparison, which show the advantage of p-ISOMAP. We also show interesting visualization applications of p-ISOMAP and demonstrate how to discover various characteristics of data through visualizations using different parameter values.

AB - One of the most widely-used nonlinear data embedding methods is ISOMAP. Based on a manifold learning framework, ISOMAP has a parameter k or ε that controls how many edges a neighborhood graph has. However, a suitable parameter value is often difficult to determine because of a time-consuming optimization process based on certain criteria, which may not be clearly justified. When ISOMAP is used to visualize data, users might want to test different parameter values in order to gain various insights about data, but such interaction between humans and such visualizations requires reasonably efficient updating, even for large-scale data. To tackle these problems, we propose an efficient updating algorithm for ISOMAP with parameter changes, called p-ISOMAP. We present not only a complexity analysis but also an empirical running time comparison, which show the advantage of p-ISOMAP. We also show interesting visualization applications of p-ISOMAP and demonstrate how to discover various characteristics of data through visualizations using different parameter values.

UR - http://www.scopus.com/inward/record.url?scp=84858052981&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84858052981&partnerID=8YFLogxK

M3 - Paper

AN - SCOPUS:84858052981

SP - 502

EP - 513

ER -