Co-Clustering of Time-Evolving News Story with Transcript and Keyframe

Author

Wu, Xiao ; Ngo, Chong-Wah ; Li, Qing

Author_Institution

Dept. of Comput. Sci., City Univ. of Hong Kong

fYear

2005

fDate

6-6 July 2005

Firstpage

117

Lastpage

120

Abstract

This paper presents techniques in clustering the same-topic news stories according to event themes. We model the relationship of stories with textual and visual concepts under the representation of bipartite graph. The textual and visual concepts are extracted respectively from speech transcripts and keyframes. Co-clustering algorithm is employed to exploit the duality of stories and textual-visual concepts based on spectral graph partitioning. Experimental results on TRECVID-2004 corpus show that the co-clustering of news stories with textual-visual concepts is significantly better than the co-clustering with either textual or visual concept alone

Keywords

feature extraction; graph theory; pattern clustering; speech processing; visual communication; TRECVID-2004 corpus; bipartite graph representation; coclustering algorithm; spectral graph partitioning; speech keyframe; speech transcript; textual-visual concept extraction; time-evolving news story; Assembly; Bipartite graph; Clustering algorithms; Computer science; Data mining; Oceans; Partitioning algorithms; Speech; Tsunami; Videos;

fLanguage

English

Publisher

ieee

Conference_Titel

Multimedia and Expo, 2005. ICME 2005. IEEE International Conference on

Conference_Location

Amsterdam

Print_ISBN

0-7803-9331-7

Type

conf

DOI

10.1109/ICME.2005.1521374

Filename

1521374