Naslov
Deliverable 4.1: Report on abstract model and P2P protocols
(Report on abstract model and P2P protocols)

Autori
Podnar Žarko, Ivana ; Rajman, Martin ; Luu, Toan ; Klemm, Fabius ; Aberer, Karl

Izvornik
Alvis Deliverable 4.1

Vrsta, podvrsta
Ostale vrste radova, izvještaj

Godina
2006

Ključne riječi
peer-to-peer overlay networks; information retrieval

Sažetak
Web search over peer-to-peer (P2P) overlay networks shows promise to enable attractive search scenarios operating at a large scale. However, the design of effective techniques for P2P indexing and retrieval raises a number of technical challenges due to potentially unscalable bandwidth consumption, and the unavailability of global document collection statistics. We report our research progress in designing and building a P2P search engine that is scalable for very large number of peers, and supports multi-term queries while providing a retrieval quality comparable to centralized solutions. The report presents a framework for full-text information retrieval in P2P overlay networks, and introduces a novel retrieval model based on highly discriminative keys. We are building a global key index in structured P2P overlays for large document collections. To cope with the problem of high bandwidth consumption, we index keys---rare and discriminative terms and term sets appearing in a restricted number of collection documents. Our indexing approach limits the size of the global index, while ensuring scalable search cost which we prove through a theoretical scalability analysis. The initial experimental results show acceptable indexing costs of the key indexing approach while the retrieval quality is comparable to standard centralized solutions with TF-IDF ranking. The architecture of our search engine is layered: On top of the P2P layer is a distributed information retrieval layer that builds a distributed index, and uses a ranking component to produce ranked answers to user queries. We present algorithms for computing the keys that are indexed by our search engine, and for maintaining a global inverted index in a completely decentralized fashion. Next, we extend the `standard' P2P layer to maintain global document collection statistics which are essential for our key-generation algorithms and ranking, and introduce special mechanisms for congestion control in P2P overlays to handle the specific indexing load. Finally, we present the initial experimental results to investigate the indexing properties of our P2P search engine.

Izvorni jezik
Engleski

Znanstvena područja
Računarstvo

Napomena
FP6 project "Alvis: Superpeer Semantic Search Engine"

POVEZANOST RADA