An SVD-based Multimodal Clustering method for Social Event Detection

Yun Ma; Qing Li; Zhenguo Yang; Zheng Lu; Haiwei Pan; Antoni B. Chan

doi:10.1109/ICDEW.2015.7129577

An SVD-based Multimodal Clustering method for Social Event Detection

Yun Ma, Qing Li, Zhenguo Yang, Zheng Lu, Haiwei Pan, Antoni B. Chan

Research output: Chapter in Book/Conference proceeding › Conference contribution › peer-review

4 Citations (Scopus)

Abstract

With the rapid development of social media sites such as Flickr, user-generated multimedia content on the Web has shown an explosive growth in recent years. Social event detection from these large multimedia collections has become one of the hottest topics in analysis of Web content. In this paper, an SVD-based Multimodal Clustering (SVDMC) algorithm is proposed to detect social events from multimodal data. SVDMC is a completely unsupervised approach aiming to take full advantage of the data at hand. Through using the binary adjacency matrix and Singular Value Decomposition (SVD), SVDMC is robust to data incompleteness for datasets in real world. Experiments conducted on the MediaEval Social Event Detection (SED) 2012 dataset demonstrate the effectiveness of the proposed method as well as discriminative power of different features.

Original language	English
Title of host publication	ICDEW 2015 - 2015 IEEE 31st International Conference on Data Engineering Workshops
Publisher	IEEE Computer Society
Pages	202-209
Number of pages	8
ISBN (Electronic)	9781479984411
DOIs	https://doi.org/10.1109/ICDEW.2015.7129577
Publication status	Published - 19 Jun 2015
Externally published	Yes
Event	2015 31st IEEE International Conference on Data Engineering Workshops, ICDEW 2015 - Seoul, Korea, Republic of Duration: 13 Apr 2015 → 17 Apr 2015

Publication series

Name	Proceedings - International Conference on Data Engineering
Volume	2015-June
ISSN (Print)	1084-4627

Conference

Conference	2015 31st IEEE International Conference on Data Engineering Workshops, ICDEW 2015
Country/Territory	Korea, Republic of
City	Seoul
Period	13/04/15 → 17/04/15

ASJC Scopus subject areas

Software
Signal Processing
Information Systems

Access to Document

10.1109/ICDEW.2015.7129577

Cite this

Ma, Y., Li, Q., Yang, Z., Lu, Z., Pan, H., & Chan, A. B. (2015). An SVD-based Multimodal Clustering method for Social Event Detection. In ICDEW 2015 - 2015 IEEE 31st International Conference on Data Engineering Workshops (pp. 202-209). Article 7129577 (Proceedings - International Conference on Data Engineering; Vol. 2015-June). IEEE Computer Society. https://doi.org/10.1109/ICDEW.2015.7129577

@inproceedings{b595c0c78e674ce0954c491e87b55661,

title = "An SVD-based Multimodal Clustering method for Social Event Detection",

abstract = "With the rapid development of social media sites such as Flickr, user-generated multimedia content on the Web has shown an explosive growth in recent years. Social event detection from these large multimedia collections has become one of the hottest topics in analysis of Web content. In this paper, an SVD-based Multimodal Clustering (SVDMC) algorithm is proposed to detect social events from multimodal data. SVDMC is a completely unsupervised approach aiming to take full advantage of the data at hand. Through using the binary adjacency matrix and Singular Value Decomposition (SVD), SVDMC is robust to data incompleteness for datasets in real world. Experiments conducted on the MediaEval Social Event Detection (SED) 2012 dataset demonstrate the effectiveness of the proposed method as well as discriminative power of different features.",

author = "Yun Ma and Qing Li and Zhenguo Yang and Zheng Lu and Haiwei Pan and Chan, {Antoni B.}",

note = "Publisher Copyright: {\textcopyright} 2015 IEEE.; 2015 31st IEEE International Conference on Data Engineering Workshops, ICDEW 2015 ; Conference date: 13-04-2015 Through 17-04-2015",

year = "2015",

month = jun,

day = "19",

doi = "10.1109/ICDEW.2015.7129577",

language = "English",

series = "Proceedings - International Conference on Data Engineering",

publisher = "IEEE Computer Society",

pages = "202--209",

booktitle = "ICDEW 2015 - 2015 IEEE 31st International Conference on Data Engineering Workshops",

address = "United States",

}

Ma, Y, Li, Q, Yang, Z, Lu, Z, Pan, H & Chan, AB 2015, An SVD-based Multimodal Clustering method for Social Event Detection. in ICDEW 2015 - 2015 IEEE 31st International Conference on Data Engineering Workshops., 7129577, Proceedings - International Conference on Data Engineering, vol. 2015-June, IEEE Computer Society, pp. 202-209, 2015 31st IEEE International Conference on Data Engineering Workshops, ICDEW 2015, Seoul, Korea, Republic of, 13/04/15. https://doi.org/10.1109/ICDEW.2015.7129577

An SVD-based Multimodal Clustering method for Social Event Detection. / Ma, Yun; Li, Qing; Yang, Zhenguo et al.
ICDEW 2015 - 2015 IEEE 31st International Conference on Data Engineering Workshops. IEEE Computer Society, 2015. p. 202-209 7129577 (Proceedings - International Conference on Data Engineering; Vol. 2015-June).

Research output: Chapter in Book/Conference proceeding › Conference contribution › peer-review

TY - GEN

T1 - An SVD-based Multimodal Clustering method for Social Event Detection

AU - Ma, Yun

AU - Li, Qing

AU - Yang, Zhenguo

AU - Lu, Zheng

AU - Pan, Haiwei

AU - Chan, Antoni B.

PY - 2015/6/19

Y1 - 2015/6/19

N2 - With the rapid development of social media sites such as Flickr, user-generated multimedia content on the Web has shown an explosive growth in recent years. Social event detection from these large multimedia collections has become one of the hottest topics in analysis of Web content. In this paper, an SVD-based Multimodal Clustering (SVDMC) algorithm is proposed to detect social events from multimodal data. SVDMC is a completely unsupervised approach aiming to take full advantage of the data at hand. Through using the binary adjacency matrix and Singular Value Decomposition (SVD), SVDMC is robust to data incompleteness for datasets in real world. Experiments conducted on the MediaEval Social Event Detection (SED) 2012 dataset demonstrate the effectiveness of the proposed method as well as discriminative power of different features.

AB - With the rapid development of social media sites such as Flickr, user-generated multimedia content on the Web has shown an explosive growth in recent years. Social event detection from these large multimedia collections has become one of the hottest topics in analysis of Web content. In this paper, an SVD-based Multimodal Clustering (SVDMC) algorithm is proposed to detect social events from multimodal data. SVDMC is a completely unsupervised approach aiming to take full advantage of the data at hand. Through using the binary adjacency matrix and Singular Value Decomposition (SVD), SVDMC is robust to data incompleteness for datasets in real world. Experiments conducted on the MediaEval Social Event Detection (SED) 2012 dataset demonstrate the effectiveness of the proposed method as well as discriminative power of different features.

UR - http://www.scopus.com/inward/record.url?scp=84944324715&partnerID=8YFLogxK

U2 - 10.1109/ICDEW.2015.7129577

DO - 10.1109/ICDEW.2015.7129577

M3 - Conference contribution

AN - SCOPUS:84944324715

T3 - Proceedings - International Conference on Data Engineering

SP - 202

EP - 209

BT - ICDEW 2015 - 2015 IEEE 31st International Conference on Data Engineering Workshops

PB - IEEE Computer Society

T2 - 2015 31st IEEE International Conference on Data Engineering Workshops, ICDEW 2015

Y2 - 13 April 2015 through 17 April 2015

ER -

An SVD-based Multimodal Clustering method for Social Event Detection

Abstract

Publication series

Conference

ASJC Scopus subject areas

Access to Document

Other files and links

Fingerprint

Cite this