Embedding heterogeneous data by preserving multiple kernels

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

1 Citation (Scopus)

Abstract

Heterogeneous data may arise in many real-life applications under different scenarios. In this paper, we formulate a general framework to address the problem of modeling heterogeneous data. Our main contribution is a novel embedding method, called multiple kernel preserving embedding (MKPE), which projects heterogeneous data into a unified embedding space by preserving crossdomain interactions and within-domain similarities simultaneously. These interactions and similarities between data points are approximated with Gaussian kernels to transfer local neighborhood information to the projected subspace. We also extend our method for out-of-sample embedding using a parametric formulation in the projection step. The performance of MKPE is illustrated on two tasks: (i) modeling biological interaction networks and (ii) cross-domain information retrieval. Empirical results of these two tasks validate the predictive performance of our algorithm.

Original languageEnglish (US)
Title of host publicationECAI 2014 - 21st European Conference on Artificial Intelligence, Including Prestigious Applications of Intelligent Systems, PAIS 2014, Proceedings
PublisherIOS Press
Pages381-386
Number of pages6
Volume263
ISBN (Print)9781614994183
DOIs
StatePublished - 2014
Event21st European Conference on Artificial Intelligence, ECAI 2014 - Prague, Czech Republic
Duration: Aug 18 2014Aug 22 2014

Publication series

NameFrontiers in Artificial Intelligence and Applications
Volume263
ISSN (Print)09226389

Other

Other21st European Conference on Artificial Intelligence, ECAI 2014
CountryCzech Republic
CityPrague
Period8/18/148/22/14

Fingerprint

Information retrieval

ASJC Scopus subject areas

  • Artificial Intelligence

Cite this

Gonen, M. (2014). Embedding heterogeneous data by preserving multiple kernels. In ECAI 2014 - 21st European Conference on Artificial Intelligence, Including Prestigious Applications of Intelligent Systems, PAIS 2014, Proceedings (Vol. 263, pp. 381-386). (Frontiers in Artificial Intelligence and Applications; Vol. 263). IOS Press. https://doi.org/10.3233/978-1-61499-419-0-381

Embedding heterogeneous data by preserving multiple kernels. / Gonen, Mehmet.

ECAI 2014 - 21st European Conference on Artificial Intelligence, Including Prestigious Applications of Intelligent Systems, PAIS 2014, Proceedings. Vol. 263 IOS Press, 2014. p. 381-386 (Frontiers in Artificial Intelligence and Applications; Vol. 263).

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

Gonen, M 2014, Embedding heterogeneous data by preserving multiple kernels. in ECAI 2014 - 21st European Conference on Artificial Intelligence, Including Prestigious Applications of Intelligent Systems, PAIS 2014, Proceedings. vol. 263, Frontiers in Artificial Intelligence and Applications, vol. 263, IOS Press, pp. 381-386, 21st European Conference on Artificial Intelligence, ECAI 2014, Prague, Czech Republic, 8/18/14. https://doi.org/10.3233/978-1-61499-419-0-381
Gonen M. Embedding heterogeneous data by preserving multiple kernels. In ECAI 2014 - 21st European Conference on Artificial Intelligence, Including Prestigious Applications of Intelligent Systems, PAIS 2014, Proceedings. Vol. 263. IOS Press. 2014. p. 381-386. (Frontiers in Artificial Intelligence and Applications). https://doi.org/10.3233/978-1-61499-419-0-381
Gonen, Mehmet. / Embedding heterogeneous data by preserving multiple kernels. ECAI 2014 - 21st European Conference on Artificial Intelligence, Including Prestigious Applications of Intelligent Systems, PAIS 2014, Proceedings. Vol. 263 IOS Press, 2014. pp. 381-386 (Frontiers in Artificial Intelligence and Applications).
@inproceedings{de709425c7b743e182c301896941ac6f,
title = "Embedding heterogeneous data by preserving multiple kernels",
abstract = "Heterogeneous data may arise in many real-life applications under different scenarios. In this paper, we formulate a general framework to address the problem of modeling heterogeneous data. Our main contribution is a novel embedding method, called multiple kernel preserving embedding (MKPE), which projects heterogeneous data into a unified embedding space by preserving crossdomain interactions and within-domain similarities simultaneously. These interactions and similarities between data points are approximated with Gaussian kernels to transfer local neighborhood information to the projected subspace. We also extend our method for out-of-sample embedding using a parametric formulation in the projection step. The performance of MKPE is illustrated on two tasks: (i) modeling biological interaction networks and (ii) cross-domain information retrieval. Empirical results of these two tasks validate the predictive performance of our algorithm.",
author = "Mehmet Gonen",
year = "2014",
doi = "10.3233/978-1-61499-419-0-381",
language = "English (US)",
isbn = "9781614994183",
volume = "263",
series = "Frontiers in Artificial Intelligence and Applications",
publisher = "IOS Press",
pages = "381--386",
booktitle = "ECAI 2014 - 21st European Conference on Artificial Intelligence, Including Prestigious Applications of Intelligent Systems, PAIS 2014, Proceedings",

}

TY - GEN

T1 - Embedding heterogeneous data by preserving multiple kernels

AU - Gonen, Mehmet

PY - 2014

Y1 - 2014

N2 - Heterogeneous data may arise in many real-life applications under different scenarios. In this paper, we formulate a general framework to address the problem of modeling heterogeneous data. Our main contribution is a novel embedding method, called multiple kernel preserving embedding (MKPE), which projects heterogeneous data into a unified embedding space by preserving crossdomain interactions and within-domain similarities simultaneously. These interactions and similarities between data points are approximated with Gaussian kernels to transfer local neighborhood information to the projected subspace. We also extend our method for out-of-sample embedding using a parametric formulation in the projection step. The performance of MKPE is illustrated on two tasks: (i) modeling biological interaction networks and (ii) cross-domain information retrieval. Empirical results of these two tasks validate the predictive performance of our algorithm.

AB - Heterogeneous data may arise in many real-life applications under different scenarios. In this paper, we formulate a general framework to address the problem of modeling heterogeneous data. Our main contribution is a novel embedding method, called multiple kernel preserving embedding (MKPE), which projects heterogeneous data into a unified embedding space by preserving crossdomain interactions and within-domain similarities simultaneously. These interactions and similarities between data points are approximated with Gaussian kernels to transfer local neighborhood information to the projected subspace. We also extend our method for out-of-sample embedding using a parametric formulation in the projection step. The performance of MKPE is illustrated on two tasks: (i) modeling biological interaction networks and (ii) cross-domain information retrieval. Empirical results of these two tasks validate the predictive performance of our algorithm.

UR - http://www.scopus.com/inward/record.url?scp=84923169054&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84923169054&partnerID=8YFLogxK

U2 - 10.3233/978-1-61499-419-0-381

DO - 10.3233/978-1-61499-419-0-381

M3 - Conference contribution

AN - SCOPUS:84923169054

SN - 9781614994183

VL - 263

T3 - Frontiers in Artificial Intelligence and Applications

SP - 381

EP - 386

BT - ECAI 2014 - 21st European Conference on Artificial Intelligence, Including Prestigious Applications of Intelligent Systems, PAIS 2014, Proceedings

PB - IOS Press

ER -