Gaussian Mixture Model-Based Subspace Modeling for Semantic Concept Retrieval

Chao Chen, Mei-Ling Shyu, Shu Ching Chen

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

Abstract

Data mining and machine learning methods have been playing an important role in searching and retrieving multimedia information from all kinds of multimedia repositories. Although some of these methods have been proven to be useful, it is still an interesting and active research area to effectively and efficiently retrieve multimedia information under difficult scenarios, i.e., detecting rare events or learning from imbalanced datasets. In this paper, we propose a novel subspace modeling framework that is able to effectively retrieve semantic concepts from highly imbalanced datasets. The proposed framework builds positive subspace models on a set of positive training sets, each of which is generated by a Gaussian Mixture Model (GMM) that partitions the data instances of a target concept (i.e., the original positive set of the target concept) into several subsets and later merges each subset with the original positive data instances. Afterwards, a joint-scoring method is proposed to fuse the final ranking scores from all such positive subspace models and the negative subspace model. Experimental results evaluated on a public-available benchmark dataset show that the proposed subspace modeling framework is able to outperform peer methods commonly used for semantic concept retrieval.

Original languageEnglish (US)
Title of host publicationProceedings - 2015 IEEE 16th International Conference on Information Reuse and Integration, IRI 2015
PublisherInstitute of Electrical and Electronics Engineers Inc.
Pages258-265
Number of pages8
ISBN (Print)9781467366564
DOIs
StatePublished - Oct 19 2015
Event16th IEEE International Conference on Information Reuse and Integration, IRI 2015 - San Francisco, United States
Duration: Aug 13 2015Aug 15 2015

Other

Other16th IEEE International Conference on Information Reuse and Integration, IRI 2015
CountryUnited States
CitySan Francisco
Period8/13/158/15/15

Fingerprint

Semantics
Electric fuses
Data mining
Learning systems
Modeling
Gaussian mixture model
Multimedia

Keywords

  • Gaussian Mixture Model (GMM)
  • Semantic Concept Retrieval
  • Subspace Modeling

ASJC Scopus subject areas

  • Information Systems
  • Information Systems and Management
  • Electrical and Electronic Engineering

Cite this

Chen, C., Shyu, M-L., & Chen, S. C. (2015). Gaussian Mixture Model-Based Subspace Modeling for Semantic Concept Retrieval. In Proceedings - 2015 IEEE 16th International Conference on Information Reuse and Integration, IRI 2015 (pp. 258-265). [7300986] Institute of Electrical and Electronics Engineers Inc.. https://doi.org/10.1109/IRI.2015.50

Gaussian Mixture Model-Based Subspace Modeling for Semantic Concept Retrieval. / Chen, Chao; Shyu, Mei-Ling; Chen, Shu Ching.

Proceedings - 2015 IEEE 16th International Conference on Information Reuse and Integration, IRI 2015. Institute of Electrical and Electronics Engineers Inc., 2015. p. 258-265 7300986.

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

Chen, C, Shyu, M-L & Chen, SC 2015, Gaussian Mixture Model-Based Subspace Modeling for Semantic Concept Retrieval. in Proceedings - 2015 IEEE 16th International Conference on Information Reuse and Integration, IRI 2015., 7300986, Institute of Electrical and Electronics Engineers Inc., pp. 258-265, 16th IEEE International Conference on Information Reuse and Integration, IRI 2015, San Francisco, United States, 8/13/15. https://doi.org/10.1109/IRI.2015.50
Chen C, Shyu M-L, Chen SC. Gaussian Mixture Model-Based Subspace Modeling for Semantic Concept Retrieval. In Proceedings - 2015 IEEE 16th International Conference on Information Reuse and Integration, IRI 2015. Institute of Electrical and Electronics Engineers Inc. 2015. p. 258-265. 7300986 https://doi.org/10.1109/IRI.2015.50
Chen, Chao ; Shyu, Mei-Ling ; Chen, Shu Ching. / Gaussian Mixture Model-Based Subspace Modeling for Semantic Concept Retrieval. Proceedings - 2015 IEEE 16th International Conference on Information Reuse and Integration, IRI 2015. Institute of Electrical and Electronics Engineers Inc., 2015. pp. 258-265
@inproceedings{5cf9c4b22fde4d58883b4522c42483df,
title = "Gaussian Mixture Model-Based Subspace Modeling for Semantic Concept Retrieval",
abstract = "Data mining and machine learning methods have been playing an important role in searching and retrieving multimedia information from all kinds of multimedia repositories. Although some of these methods have been proven to be useful, it is still an interesting and active research area to effectively and efficiently retrieve multimedia information under difficult scenarios, i.e., detecting rare events or learning from imbalanced datasets. In this paper, we propose a novel subspace modeling framework that is able to effectively retrieve semantic concepts from highly imbalanced datasets. The proposed framework builds positive subspace models on a set of positive training sets, each of which is generated by a Gaussian Mixture Model (GMM) that partitions the data instances of a target concept (i.e., the original positive set of the target concept) into several subsets and later merges each subset with the original positive data instances. Afterwards, a joint-scoring method is proposed to fuse the final ranking scores from all such positive subspace models and the negative subspace model. Experimental results evaluated on a public-available benchmark dataset show that the proposed subspace modeling framework is able to outperform peer methods commonly used for semantic concept retrieval.",
keywords = "Gaussian Mixture Model (GMM), Semantic Concept Retrieval, Subspace Modeling",
author = "Chao Chen and Mei-Ling Shyu and Chen, {Shu Ching}",
year = "2015",
month = "10",
day = "19",
doi = "10.1109/IRI.2015.50",
language = "English (US)",
isbn = "9781467366564",
pages = "258--265",
booktitle = "Proceedings - 2015 IEEE 16th International Conference on Information Reuse and Integration, IRI 2015",
publisher = "Institute of Electrical and Electronics Engineers Inc.",

}

TY - GEN

T1 - Gaussian Mixture Model-Based Subspace Modeling for Semantic Concept Retrieval

AU - Chen, Chao

AU - Shyu, Mei-Ling

AU - Chen, Shu Ching

PY - 2015/10/19

Y1 - 2015/10/19

N2 - Data mining and machine learning methods have been playing an important role in searching and retrieving multimedia information from all kinds of multimedia repositories. Although some of these methods have been proven to be useful, it is still an interesting and active research area to effectively and efficiently retrieve multimedia information under difficult scenarios, i.e., detecting rare events or learning from imbalanced datasets. In this paper, we propose a novel subspace modeling framework that is able to effectively retrieve semantic concepts from highly imbalanced datasets. The proposed framework builds positive subspace models on a set of positive training sets, each of which is generated by a Gaussian Mixture Model (GMM) that partitions the data instances of a target concept (i.e., the original positive set of the target concept) into several subsets and later merges each subset with the original positive data instances. Afterwards, a joint-scoring method is proposed to fuse the final ranking scores from all such positive subspace models and the negative subspace model. Experimental results evaluated on a public-available benchmark dataset show that the proposed subspace modeling framework is able to outperform peer methods commonly used for semantic concept retrieval.

AB - Data mining and machine learning methods have been playing an important role in searching and retrieving multimedia information from all kinds of multimedia repositories. Although some of these methods have been proven to be useful, it is still an interesting and active research area to effectively and efficiently retrieve multimedia information under difficult scenarios, i.e., detecting rare events or learning from imbalanced datasets. In this paper, we propose a novel subspace modeling framework that is able to effectively retrieve semantic concepts from highly imbalanced datasets. The proposed framework builds positive subspace models on a set of positive training sets, each of which is generated by a Gaussian Mixture Model (GMM) that partitions the data instances of a target concept (i.e., the original positive set of the target concept) into several subsets and later merges each subset with the original positive data instances. Afterwards, a joint-scoring method is proposed to fuse the final ranking scores from all such positive subspace models and the negative subspace model. Experimental results evaluated on a public-available benchmark dataset show that the proposed subspace modeling framework is able to outperform peer methods commonly used for semantic concept retrieval.

KW - Gaussian Mixture Model (GMM)

KW - Semantic Concept Retrieval

KW - Subspace Modeling

UR - http://www.scopus.com/inward/record.url?scp=84959081199&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84959081199&partnerID=8YFLogxK

U2 - 10.1109/IRI.2015.50

DO - 10.1109/IRI.2015.50

M3 - Conference contribution

AN - SCOPUS:84959081199

SN - 9781467366564

SP - 258

EP - 265

BT - Proceedings - 2015 IEEE 16th International Conference on Information Reuse and Integration, IRI 2015

PB - Institute of Electrical and Electronics Engineers Inc.

ER -