TY - GEN
T1 - Hierarchical clustering of WWW image search results using visual, textual and link information
AU - Cai, Deng
AU - He, Xiaofei
AU - Li, Zhiwei
AU - Ma, Wei-Ying
AU - Wen, Ji-Rong
N1 - Publication details (e.g. title, author(s), publication statuses and dates) are captured on an “AS IS” and “AS AVAILABLE” basis at the time of record harvesting from the data source. Suggestions for further amendments or supplementary information can be sent to [email protected].
PY - 2004
Y1 - 2004
N2 - We consider the problem of clustering Web image search results. Generally, the image search results returned by an image search engine contain multiple topics. Organizing the results into different semantic clusters facilitates users' browsing. In this paper, we propose a hierarchical clustering method using visual, textual and link analysis. By using a vision-based page segmentation algorithm, a web page is partitioned into blocks, and the textual and link information of an image can be accurately extracted from the block containing that image. By using block-level link analysis techniques, an image graph can be constructed. We then apply spectral techniques to find a Euclidean embedding of the images which respects the graph structure, Thus for each image, we have three kinds of representations, i.e. visual feature based representation, textual feature based representation and graph based representation. Using spectral clustering techniques, we can cluster the search results into different semantic clusters. An image search example illustrates the potential of these techniques.
AB - We consider the problem of clustering Web image search results. Generally, the image search results returned by an image search engine contain multiple topics. Organizing the results into different semantic clusters facilitates users' browsing. In this paper, we propose a hierarchical clustering method using visual, textual and link analysis. By using a vision-based page segmentation algorithm, a web page is partitioned into blocks, and the textual and link information of an image can be accurately extracted from the block containing that image. By using block-level link analysis techniques, an image graph can be constructed. We then apply spectral techniques to find a Euclidean embedding of the images which respects the graph structure, Thus for each image, we have three kinds of representations, i.e. visual feature based representation, textual feature based representation and graph based representation. Using spectral clustering techniques, we can cluster the search results into different semantic clusters. An image search example illustrates the potential of these techniques.
KW - Graph Model
KW - Image Clustering
KW - Link Analysis
KW - Search Result Organization
KW - Spectral Analysis
KW - Vision Based Page Segmentation
KW - Web Image Search
UR - http://www.scopus.com/inward/record.url?scp=13444282389&partnerID=8YFLogxK
UR - https://www.scopus.com/record/pubmetrics.uri?eid=2-s2.0-13444282389&origin=recordpage
U2 - 10.1145/1027527.1027747
DO - 10.1145/1027527.1027747
M3 - RGC 32 - Refereed conference paper (with host publication)
SN - 1581138938
SN - 9781581138931
T3 - ACM Multimedia 2004 - proceedings of the 12th ACM International Conference on Multimedia
SP - 952
EP - 959
BT - ACM Multimedia 2004 - proceedings of the 12th ACM International Conference on Multimedia
PB - Association for Computing Machinery
T2 - ACM Multimedia 2004 - proceedings of the 12th ACM International Conference on Multimedia
Y2 - 10 October 2004 through 16 October 2004
ER -