Heterogeneous data fusion via space alignment using nonmetric multidimensional scaling

Jaegul Choo, Shawn Bohn, Grant C. Nakamura, Amanda M. White, Haesun Park

Research output: Chapter in Book/Report/Conference proceedingConference contribution

16 Citations (Scopus)

Abstract

Heterogeneous data sets are typically represented in different feature spaces, making it difficult to analyze relationships spanning different data sets even when they are semantically related. Data fusion via space alignment can remedy this task by integrating multiple data sets lying in different spaces into one common space. Given a set of reference correspondence data that share the same semantic meaning across different spaces, space alignment attempts to place the corresponding reference data as close together as possible, and accordingly, the entire data are aligned in a common space. Space alignment involves optimizing two potentially conflicting criteria: minimum deformation of the original relationships and maximum alignment between the different spaces. To solve this problem, we provide a novel graph embedding framework for space alignment, which converts each data set into a graph and assigns zero distance between reference correspondence pairs resulting in a single graph. We propose a graph embedding method for fusion based on nonmetric multidimensional scaling (MDS). Its criteria using the rank order rather than the distance allows nonmetric MDS to effectively handle both deformation and alignment. Experiments using parallel data sets demonstrate that our approach works well in comparison to existing methods such as constrained Laplacian eigenmaps, Procrustes analysis, and tensor decomposition. We also present standard cross-domain information retrieval tests as well as interesting visualization examples using space alignment.

Original languageEnglish
Title of host publicationProceedings of the 12th SIAM International Conference on Data Mining, SDM 2012
Pages177-188
Number of pages12
Publication statusPublished - 2012 Dec 1
Externally publishedYes
Event12th SIAM International Conference on Data Mining, SDM 2012 - Anaheim, CA, United States
Duration: 2012 Apr 262012 Apr 28

Publication series

NameProceedings of the 12th SIAM International Conference on Data Mining, SDM 2012

Conference

Conference12th SIAM International Conference on Data Mining, SDM 2012
CountryUnited States
CityAnaheim, CA
Period12/4/2612/4/28

Fingerprint

Data fusion
Information retrieval
Tensors
Visualization
Semantics
Decomposition

ASJC Scopus subject areas

  • Computer Science Applications

Cite this

Choo, J., Bohn, S., Nakamura, G. C., White, A. M., & Park, H. (2012). Heterogeneous data fusion via space alignment using nonmetric multidimensional scaling. In Proceedings of the 12th SIAM International Conference on Data Mining, SDM 2012 (pp. 177-188). (Proceedings of the 12th SIAM International Conference on Data Mining, SDM 2012).

Heterogeneous data fusion via space alignment using nonmetric multidimensional scaling. / Choo, Jaegul; Bohn, Shawn; Nakamura, Grant C.; White, Amanda M.; Park, Haesun.

Proceedings of the 12th SIAM International Conference on Data Mining, SDM 2012. 2012. p. 177-188 (Proceedings of the 12th SIAM International Conference on Data Mining, SDM 2012).

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Choo, J, Bohn, S, Nakamura, GC, White, AM & Park, H 2012, Heterogeneous data fusion via space alignment using nonmetric multidimensional scaling. in Proceedings of the 12th SIAM International Conference on Data Mining, SDM 2012. Proceedings of the 12th SIAM International Conference on Data Mining, SDM 2012, pp. 177-188, 12th SIAM International Conference on Data Mining, SDM 2012, Anaheim, CA, United States, 12/4/26.
Choo J, Bohn S, Nakamura GC, White AM, Park H. Heterogeneous data fusion via space alignment using nonmetric multidimensional scaling. In Proceedings of the 12th SIAM International Conference on Data Mining, SDM 2012. 2012. p. 177-188. (Proceedings of the 12th SIAM International Conference on Data Mining, SDM 2012).
Choo, Jaegul ; Bohn, Shawn ; Nakamura, Grant C. ; White, Amanda M. ; Park, Haesun. / Heterogeneous data fusion via space alignment using nonmetric multidimensional scaling. Proceedings of the 12th SIAM International Conference on Data Mining, SDM 2012. 2012. pp. 177-188 (Proceedings of the 12th SIAM International Conference on Data Mining, SDM 2012).
@inproceedings{bd1e0efd35c84a179ab04152fa3142bc,
title = "Heterogeneous data fusion via space alignment using nonmetric multidimensional scaling",
abstract = "Heterogeneous data sets are typically represented in different feature spaces, making it difficult to analyze relationships spanning different data sets even when they are semantically related. Data fusion via space alignment can remedy this task by integrating multiple data sets lying in different spaces into one common space. Given a set of reference correspondence data that share the same semantic meaning across different spaces, space alignment attempts to place the corresponding reference data as close together as possible, and accordingly, the entire data are aligned in a common space. Space alignment involves optimizing two potentially conflicting criteria: minimum deformation of the original relationships and maximum alignment between the different spaces. To solve this problem, we provide a novel graph embedding framework for space alignment, which converts each data set into a graph and assigns zero distance between reference correspondence pairs resulting in a single graph. We propose a graph embedding method for fusion based on nonmetric multidimensional scaling (MDS). Its criteria using the rank order rather than the distance allows nonmetric MDS to effectively handle both deformation and alignment. Experiments using parallel data sets demonstrate that our approach works well in comparison to existing methods such as constrained Laplacian eigenmaps, Procrustes analysis, and tensor decomposition. We also present standard cross-domain information retrieval tests as well as interesting visualization examples using space alignment.",
author = "Jaegul Choo and Shawn Bohn and Nakamura, {Grant C.} and White, {Amanda M.} and Haesun Park",
year = "2012",
month = "12",
day = "1",
language = "English",
isbn = "9781611972320",
series = "Proceedings of the 12th SIAM International Conference on Data Mining, SDM 2012",
pages = "177--188",
booktitle = "Proceedings of the 12th SIAM International Conference on Data Mining, SDM 2012",

}

TY - GEN

T1 - Heterogeneous data fusion via space alignment using nonmetric multidimensional scaling

AU - Choo, Jaegul

AU - Bohn, Shawn

AU - Nakamura, Grant C.

AU - White, Amanda M.

AU - Park, Haesun

PY - 2012/12/1

Y1 - 2012/12/1

N2 - Heterogeneous data sets are typically represented in different feature spaces, making it difficult to analyze relationships spanning different data sets even when they are semantically related. Data fusion via space alignment can remedy this task by integrating multiple data sets lying in different spaces into one common space. Given a set of reference correspondence data that share the same semantic meaning across different spaces, space alignment attempts to place the corresponding reference data as close together as possible, and accordingly, the entire data are aligned in a common space. Space alignment involves optimizing two potentially conflicting criteria: minimum deformation of the original relationships and maximum alignment between the different spaces. To solve this problem, we provide a novel graph embedding framework for space alignment, which converts each data set into a graph and assigns zero distance between reference correspondence pairs resulting in a single graph. We propose a graph embedding method for fusion based on nonmetric multidimensional scaling (MDS). Its criteria using the rank order rather than the distance allows nonmetric MDS to effectively handle both deformation and alignment. Experiments using parallel data sets demonstrate that our approach works well in comparison to existing methods such as constrained Laplacian eigenmaps, Procrustes analysis, and tensor decomposition. We also present standard cross-domain information retrieval tests as well as interesting visualization examples using space alignment.

AB - Heterogeneous data sets are typically represented in different feature spaces, making it difficult to analyze relationships spanning different data sets even when they are semantically related. Data fusion via space alignment can remedy this task by integrating multiple data sets lying in different spaces into one common space. Given a set of reference correspondence data that share the same semantic meaning across different spaces, space alignment attempts to place the corresponding reference data as close together as possible, and accordingly, the entire data are aligned in a common space. Space alignment involves optimizing two potentially conflicting criteria: minimum deformation of the original relationships and maximum alignment between the different spaces. To solve this problem, we provide a novel graph embedding framework for space alignment, which converts each data set into a graph and assigns zero distance between reference correspondence pairs resulting in a single graph. We propose a graph embedding method for fusion based on nonmetric multidimensional scaling (MDS). Its criteria using the rank order rather than the distance allows nonmetric MDS to effectively handle both deformation and alignment. Experiments using parallel data sets demonstrate that our approach works well in comparison to existing methods such as constrained Laplacian eigenmaps, Procrustes analysis, and tensor decomposition. We also present standard cross-domain information retrieval tests as well as interesting visualization examples using space alignment.

UR - http://www.scopus.com/inward/record.url?scp=84875848281&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84875848281&partnerID=8YFLogxK

M3 - Conference contribution

AN - SCOPUS:84875848281

SN - 9781611972320

T3 - Proceedings of the 12th SIAM International Conference on Data Mining, SDM 2012

SP - 177

EP - 188

BT - Proceedings of the 12th SIAM International Conference on Data Mining, SDM 2012

ER -