DMCA
A scalable semantic indexing framework for peer-to-peer information retrieval (2005)
| Venue: | In SIGIR 2005 workshop: Heterogeneous and Distributed Information Retrieval |
| Citations: | 6 - 0 self |
BibTeX
@INPROCEEDINGS{Chen05ascalable,
author = {Yan Chen},
title = {A scalable semantic indexing framework for peer-to-peer information retrieval},
booktitle = {In SIGIR 2005 workshop: Heterogeneous and Distributed Information Retrieval},
year = {2005}
}
Share
 |
 |
 |
 |
||
OpenURL
Abstract
The exponential growth of data demands scalable and adaptable infrastructures for indexing and searching a huge amount of data sources with high accuracy and efficiency. Existing centralized search engines are not scalable and suffer from single-point-offailures. The recent work on P2P index construction partitions the document vectors either randomly or statically, making it difficult to tradeoff between search efficiency and accuracy. In this position paper, we propose a peer-to-peer (P2P) IR framework (termed as P2PIR) that leverages a novel two-phase distributed semantic indexing on top of distributed hash tables (DHT). The distributed semantic clustering of P2PIR leads to good semantic locality on index placement so that the indices of similar documents are placed together or near to each other. The semantic locality enables smoother tradeoff between search accuracy and efficiency, as well as incremental adaptation to document and semantics changes. In addition, P2PIR allows for sophisticated retrieval techniques, e.g., query refinement, feedback and personalized search for better usability. A prototype of P2PIR is currently under development, which can be applied for general web retrieval and domain-specific applications such as a distributed electric medical records system. 1.
Keyphrases
scalable semantic indexing framework peer-to-peer information retrieval good semantic locality sophisticated retrieval technique document vector semantics change huge amount semantic indexing similar document p2p index construction adaptable infrastructure search efficiency query refinement position paper data source distributed electric medical record system exponential growth semantic locality search engine domain-specific application index placement incremental adaptation novel two-phase data demand high accuracy recent work search accuracy distributed semantic clustering distributed hash table general web retrieval ir framework
