Multiple kernel learning clustering with an application to malware

Blake Anderson, Curtis Storlie, Terran Lane

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

7 Citations (Scopus)

Abstract

With the increasing prevalence of richer, more complex data sources, learning with multiple views is becoming more widespread. Multiple kernel learning (MKL) has been developed to address this problem, but in general, the solutions provided by traditional MKL are restricted to a classification objective function. In this work, we develop a novel multiple kernel learning algorithm that is based on a spectral clustering objective function which is able to find an optimal kernel weight vector for the clustering problem. We go on to show how this optimization problem can be cast as a semidefinite program and efficiently solved using off-the-shelf interior point methods.

Original languageEnglish (US)
Title of host publicationProceedings - 12th IEEE International Conference on Data Mining, ICDM 2012
Pages804-809
Number of pages6
DOIs
StatePublished - 2012
Externally publishedYes
Event12th IEEE International Conference on Data Mining, ICDM 2012 - Brussels, Belgium
Duration: Dec 10 2012Dec 13 2012

Other

Other12th IEEE International Conference on Data Mining, ICDM 2012
CountryBelgium
CityBrussels
Period12/10/1212/13/12

Fingerprint

Learning algorithms
Malware

Keywords

  • Malware
  • Multiple kernel learning
  • Semidefinite programming
  • Spectral clustering

ASJC Scopus subject areas

  • Engineering(all)

Cite this

Anderson, B., Storlie, C., & Lane, T. (2012). Multiple kernel learning clustering with an application to malware. In Proceedings - 12th IEEE International Conference on Data Mining, ICDM 2012 (pp. 804-809). [6413849] https://doi.org/10.1109/ICDM.2012.75

Multiple kernel learning clustering with an application to malware. / Anderson, Blake; Storlie, Curtis; Lane, Terran.

Proceedings - 12th IEEE International Conference on Data Mining, ICDM 2012. 2012. p. 804-809 6413849.

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

Anderson, B, Storlie, C & Lane, T 2012, Multiple kernel learning clustering with an application to malware. in Proceedings - 12th IEEE International Conference on Data Mining, ICDM 2012., 6413849, pp. 804-809, 12th IEEE International Conference on Data Mining, ICDM 2012, Brussels, Belgium, 12/10/12. https://doi.org/10.1109/ICDM.2012.75
Anderson B, Storlie C, Lane T. Multiple kernel learning clustering with an application to malware. In Proceedings - 12th IEEE International Conference on Data Mining, ICDM 2012. 2012. p. 804-809. 6413849 https://doi.org/10.1109/ICDM.2012.75
Anderson, Blake ; Storlie, Curtis ; Lane, Terran. / Multiple kernel learning clustering with an application to malware. Proceedings - 12th IEEE International Conference on Data Mining, ICDM 2012. 2012. pp. 804-809
@inproceedings{6eead2fd17964a68b8dd90c349b7ccf5,
title = "Multiple kernel learning clustering with an application to malware",
abstract = "With the increasing prevalence of richer, more complex data sources, learning with multiple views is becoming more widespread. Multiple kernel learning (MKL) has been developed to address this problem, but in general, the solutions provided by traditional MKL are restricted to a classification objective function. In this work, we develop a novel multiple kernel learning algorithm that is based on a spectral clustering objective function which is able to find an optimal kernel weight vector for the clustering problem. We go on to show how this optimization problem can be cast as a semidefinite program and efficiently solved using off-the-shelf interior point methods.",
keywords = "Malware, Multiple kernel learning, Semidefinite programming, Spectral clustering",
author = "Blake Anderson and Curtis Storlie and Terran Lane",
year = "2012",
doi = "10.1109/ICDM.2012.75",
language = "English (US)",
isbn = "9780769549057",
pages = "804--809",
booktitle = "Proceedings - 12th IEEE International Conference on Data Mining, ICDM 2012",

}

TY - GEN

T1 - Multiple kernel learning clustering with an application to malware

AU - Anderson, Blake

AU - Storlie, Curtis

AU - Lane, Terran

PY - 2012

Y1 - 2012

N2 - With the increasing prevalence of richer, more complex data sources, learning with multiple views is becoming more widespread. Multiple kernel learning (MKL) has been developed to address this problem, but in general, the solutions provided by traditional MKL are restricted to a classification objective function. In this work, we develop a novel multiple kernel learning algorithm that is based on a spectral clustering objective function which is able to find an optimal kernel weight vector for the clustering problem. We go on to show how this optimization problem can be cast as a semidefinite program and efficiently solved using off-the-shelf interior point methods.

AB - With the increasing prevalence of richer, more complex data sources, learning with multiple views is becoming more widespread. Multiple kernel learning (MKL) has been developed to address this problem, but in general, the solutions provided by traditional MKL are restricted to a classification objective function. In this work, we develop a novel multiple kernel learning algorithm that is based on a spectral clustering objective function which is able to find an optimal kernel weight vector for the clustering problem. We go on to show how this optimization problem can be cast as a semidefinite program and efficiently solved using off-the-shelf interior point methods.

KW - Malware

KW - Multiple kernel learning

KW - Semidefinite programming

KW - Spectral clustering

UR - http://www.scopus.com/inward/record.url?scp=84874033284&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84874033284&partnerID=8YFLogxK

U2 - 10.1109/ICDM.2012.75

DO - 10.1109/ICDM.2012.75

M3 - Conference contribution

SN - 9780769549057

SP - 804

EP - 809

BT - Proceedings - 12th IEEE International Conference on Data Mining, ICDM 2012

ER -