A training algorithm for optimal margin classifiers
COLT '92 Proceedings of the fifth annual workshop on Computational learning theory
Improving text retrieval for the routing problem using latent semantic indexing
SIGIR '94 Proceedings of the 17th annual international ACM SIGIR conference on Research and development in information retrieval
Chord: A scalable peer-to-peer lookup service for internet applications
Proceedings of the 2001 conference on Applications, technologies, architectures, and protocols for computer communications
A scalable content-addressable network
Proceedings of the 2001 conference on Applications, technologies, architectures, and protocols for computer communications
A Tutorial on Support Vector Machines for Pattern Recognition
Data Mining and Knowledge Discovery
Peer-to-peer information retrieval using self-organizing semantic overlay networks
Proceedings of the 2003 conference on Applications, technologies, architectures, and protocols for computer communications
Remindin': semantic query routing in peer-to-peer networks based on social metaphors
Proceedings of the 13th international conference on World Wide Web
On scaling latent semantic indexing for large peer-to-peer systems
Proceedings of the 27th annual international ACM SIGIR conference on Research and development in information retrieval
Relation discovery from web data for competency management
Web Intelligence and Agent Systems
SemreX: a semantic peer-to-peer system for literature documents retrieval
ASWC'06 Proceedings of the First Asian conference on The Semantic Web
Hi-index | 0.00 |
A fundamental problem in peer-to-peer networks is how to locate appropriate peers efficiently to answer a specific query request. This paper proposes a model in which semantically similar peers form a semantic overlay network and a query can be routed or forwarded to appropriate peers instead of broadcasting or random selection. We apply Latent Semantic Indexing (LSI) in information retrieval to reveal semantic subspaces of feature spaces from documents stored on peers. After producing semantic vectors through LSI, we train a support vector machine (SVM) to classify the peers into different categories based on the extracted vectors. Peers with close categories are defined as semantic similarity and form a semantic overlay. Experimental results show the model is efficient and performs better than other non-semantic retrieval models with respect to accuracy. In addition, our approach improves the recall rate nearly 100% while reducing message traffic dramatically compared with Gnutella.