|
Published Articles >> Table of Contents >> Abstract
Eighth IEEE International Symposium on Multimedia (ISM'06)
pp. 823-830
Clustering-Based Source Selection for Efficient Image Retrieval in Peer-to-Peer Networks
Martin Eisenhardt, University of Bamberg, Germany
Wolfgang Muller, University of Bamberg, Germany
Andreas Henrich, University of Bamberg, Germany
Daniel Blank, University of Bamberg, Germany
Soufyane El Allali, University of Bamberg, Germany
Full Article Text:
 
DOI Bookmark: http://doi.ieeecomputersociety.org/10.1109/ISM.2006.47
Send link to a friend
| Abstract |
|
In peer-to-peer (P2P) networks, computers with
equal rights form a logical (overlay) network in order
to provide a common service that lies beyond the ca-
pacity of every single participant. Efficient similarity
search is generally recognized as a frontier in research
about P2P systems. One way to address it is using data
source selection based approaches where peers summa-
rize the data they contribute to the network, generat-
ing typically one summary per peer. When process-
ing queries, these summaries are used to choose the
peers (data sources) that are most likely to contribute
to the query result. Only those data sources are con-
tacted. There are two main contributions of this paper.
We extend earlier work, adding a data source selec-
tion method for high-dimensional vector data, compar-
ing different peer ranking schemes. More importantly,
we present a method that uses progressive stepwise data
exchange between peers to better each peer’s summary
and therefore improve the system’s performance.
|
 Additional Information
|
Citation:
Martin Eisenhardt, Wolfgang Muller, Andreas Henrich, Daniel Blank, Soufyane El Allali,
"Clustering-Based Source Selection for Efficient Image Retrieval in Peer-to-Peer Networks,"
ism,
pp. 823-830,
Eighth IEEE International Symposium on Multimedia (ISM'06),
2006
|
|
