Please use this identifier to cite or link to this item: https://doi.org/10.1007/978-3-642-03869-3_48
Title: SiMPSON: Efficient similarity search in metric spaces over P2P structured overlay networks
Authors: Vu, Q.H.
Lupu, M.
Wu, S. 
Issue Date: 2009
Source: Vu, Q.H.,Lupu, M.,Wu, S. (2009). SiMPSON: Efficient similarity search in metric spaces over P2P structured overlay networks. Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics) 5704 LNCS : 498-510. ScholarBank@NUS Repository. https://doi.org/10.1007/978-3-642-03869-3_48
Abstract: Similarity search in metric spaces over centralized systems has been significantly studied in the database research community. However, not so much work has been done in the context of P2P networks. This paper introduces SiMPSON: a P2P system supporting similarity search in metric spaces. The aim is to answer queries faster and using less resources than existing systems. For this, each peer first clusters its own data using any off-the-shelf clustering algorithms. Then, the resulting clusters are mapped to one-dimensional values. Finally, these one-dimensional values are indexed into a structured P2P overlay. Our method slightly increases the indexing overhead, but allows us to greatly reduce the number of peers and messages involved in query processing: we trade a small amount of overhead in the data publishing process for a substantial reduction of costs in the querying phase. Based on this architecture, we propose algorithms for processing range and kNN queries. Extensive experimental results validate the claims of efficiency and effectiveness of SiMPSON. © 2009 Springer.
Source Title: Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
URI: http://scholarbank.nus.edu.sg/handle/10635/41146
ISBN: 3642038689
ISSN: 03029743
DOI: 10.1007/978-3-642-03869-3_48
Appears in Collections:Staff Publications

Show full item record
Files in This Item:
There are no files associated with this item.

SCOPUSTM   
Citations

7
checked on Dec 13, 2017

Page view(s)

58
checked on Dec 9, 2017

Google ScholarTM

Check

Altmetric


Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.