Correlation-based interestingness measure for video semantic concept detection

Lin Lin, Mei-Ling Shyu, Shu Ching Chen

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

3 Citations (Scopus)

Abstract

The technique of performing classification using association rule mining (ARM) has been adopted to bridge the multimedia semantic gap between low-level features and high-level concepts of interest, taking advantages of both classification and association rule mining. One of the most important research approaches in ARM is to investigate the interestingness measure which plays a key role in association rule discovery stage and rule selection stage. In this paper, a new correlation-based interestingness measure that is used at both stages is proposed. The association rules are generated by a novel interestingness measure obtained from applying multiple correspondence analysis (MCA) to explore the correlation between two feature-value pairs and concept classes. Then the correlation-based interestingness measure is reused and aggregated with the inter-similarity and intra-similarity values to rank the final rule set for classification. Detecting the concepts from the benchmark data provided by the TRECVID project, we have shown that our proposed framework achieves higher accuracy than the classifiers that are commonly applied to multimedia retrieval.

Original languageEnglish
Title of host publication2009 IEEE International Conference on Information Reuse and Integration, IRI 2009
Pages120-125
Number of pages6
DOIs
StatePublished - Nov 17 2009
Event2009 IEEE International Conference on Information Reuse and Integration, IRI 2009 - Las Vegas, NV, United States
Duration: Aug 10 2009Aug 12 2009

Other

Other2009 IEEE International Conference on Information Reuse and Integration, IRI 2009
CountryUnited States
CityLas Vegas, NV
Period8/10/098/12/09

Fingerprint

Association rules
Semantics
Classifiers

Keywords

  • Interestingness measure
  • Multiple correspondence analysis
  • Semantic concept detection

ASJC Scopus subject areas

  • Computer Networks and Communications
  • Computer Science Applications
  • Information Systems
  • Software

Cite this

Lin, L., Shyu, M-L., & Chen, S. C. (2009). Correlation-based interestingness measure for video semantic concept detection. In 2009 IEEE International Conference on Information Reuse and Integration, IRI 2009 (pp. 120-125). [5211537] https://doi.org/10.1109/IRI.2009.5211537

Correlation-based interestingness measure for video semantic concept detection. / Lin, Lin; Shyu, Mei-Ling; Chen, Shu Ching.

2009 IEEE International Conference on Information Reuse and Integration, IRI 2009. 2009. p. 120-125 5211537.

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

Lin, L, Shyu, M-L & Chen, SC 2009, Correlation-based interestingness measure for video semantic concept detection. in 2009 IEEE International Conference on Information Reuse and Integration, IRI 2009., 5211537, pp. 120-125, 2009 IEEE International Conference on Information Reuse and Integration, IRI 2009, Las Vegas, NV, United States, 8/10/09. https://doi.org/10.1109/IRI.2009.5211537
Lin L, Shyu M-L, Chen SC. Correlation-based interestingness measure for video semantic concept detection. In 2009 IEEE International Conference on Information Reuse and Integration, IRI 2009. 2009. p. 120-125. 5211537 https://doi.org/10.1109/IRI.2009.5211537
Lin, Lin ; Shyu, Mei-Ling ; Chen, Shu Ching. / Correlation-based interestingness measure for video semantic concept detection. 2009 IEEE International Conference on Information Reuse and Integration, IRI 2009. 2009. pp. 120-125
@inproceedings{2a2f20c432a84cbe8a6a870495aa33fd,
title = "Correlation-based interestingness measure for video semantic concept detection",
abstract = "The technique of performing classification using association rule mining (ARM) has been adopted to bridge the multimedia semantic gap between low-level features and high-level concepts of interest, taking advantages of both classification and association rule mining. One of the most important research approaches in ARM is to investigate the interestingness measure which plays a key role in association rule discovery stage and rule selection stage. In this paper, a new correlation-based interestingness measure that is used at both stages is proposed. The association rules are generated by a novel interestingness measure obtained from applying multiple correspondence analysis (MCA) to explore the correlation between two feature-value pairs and concept classes. Then the correlation-based interestingness measure is reused and aggregated with the inter-similarity and intra-similarity values to rank the final rule set for classification. Detecting the concepts from the benchmark data provided by the TRECVID project, we have shown that our proposed framework achieves higher accuracy than the classifiers that are commonly applied to multimedia retrieval.",
keywords = "Interestingness measure, Multiple correspondence analysis, Semantic concept detection",
author = "Lin Lin and Mei-Ling Shyu and Chen, {Shu Ching}",
year = "2009",
month = "11",
day = "17",
doi = "10.1109/IRI.2009.5211537",
language = "English",
isbn = "9781424441167",
pages = "120--125",
booktitle = "2009 IEEE International Conference on Information Reuse and Integration, IRI 2009",

}

TY - GEN

T1 - Correlation-based interestingness measure for video semantic concept detection

AU - Lin, Lin

AU - Shyu, Mei-Ling

AU - Chen, Shu Ching

PY - 2009/11/17

Y1 - 2009/11/17

N2 - The technique of performing classification using association rule mining (ARM) has been adopted to bridge the multimedia semantic gap between low-level features and high-level concepts of interest, taking advantages of both classification and association rule mining. One of the most important research approaches in ARM is to investigate the interestingness measure which plays a key role in association rule discovery stage and rule selection stage. In this paper, a new correlation-based interestingness measure that is used at both stages is proposed. The association rules are generated by a novel interestingness measure obtained from applying multiple correspondence analysis (MCA) to explore the correlation between two feature-value pairs and concept classes. Then the correlation-based interestingness measure is reused and aggregated with the inter-similarity and intra-similarity values to rank the final rule set for classification. Detecting the concepts from the benchmark data provided by the TRECVID project, we have shown that our proposed framework achieves higher accuracy than the classifiers that are commonly applied to multimedia retrieval.

AB - The technique of performing classification using association rule mining (ARM) has been adopted to bridge the multimedia semantic gap between low-level features and high-level concepts of interest, taking advantages of both classification and association rule mining. One of the most important research approaches in ARM is to investigate the interestingness measure which plays a key role in association rule discovery stage and rule selection stage. In this paper, a new correlation-based interestingness measure that is used at both stages is proposed. The association rules are generated by a novel interestingness measure obtained from applying multiple correspondence analysis (MCA) to explore the correlation between two feature-value pairs and concept classes. Then the correlation-based interestingness measure is reused and aggregated with the inter-similarity and intra-similarity values to rank the final rule set for classification. Detecting the concepts from the benchmark data provided by the TRECVID project, we have shown that our proposed framework achieves higher accuracy than the classifiers that are commonly applied to multimedia retrieval.

KW - Interestingness measure

KW - Multiple correspondence analysis

KW - Semantic concept detection

UR - http://www.scopus.com/inward/record.url?scp=70449371205&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=70449371205&partnerID=8YFLogxK

U2 - 10.1109/IRI.2009.5211537

DO - 10.1109/IRI.2009.5211537

M3 - Conference contribution

SN - 9781424441167

SP - 120

EP - 125

BT - 2009 IEEE International Conference on Information Reuse and Integration, IRI 2009

ER -