P2Prec: A P2P Recommendation System for Large-Scale Data Sharing

1263's picture
Journal Title, Volume, Page: 
Transactions on Large-Scale Data- and Knowledge-Centered Systems III Lecture Notes in Computer Science Volume 6790, 2011, pp 87-116
Year of Publication: 
2011
Authors: 
Fady Draidi
INRIA & LIRMM, Montpellier, France
Esther Pacitti
INRIA & LIRMM, Montpellier, France
Bettina Kemme
McGill University, Montreal, Canada
Preferred Abstract (Original): 
This paper proposes P2Prec, a P2P recommendation overlay that facilitates document sharing for on-line communities. Given a query, the goal of P2PRec is to find relevant peers that can recommend documents that are relevant for the query and are of high quality. A document is relevant to a query if it covers the same topics. It is of high quality if relevant peers have rated it highly. P2PRec finds relevant peers through a variety of mechanisms including advanced content-based and collaborative filtering. The topics each peer is interested in are automatically calculated by analyzing the documents the peer holds. Peers become relevant for a topic if they hold a certain number of highly rated documents on this topic. To efficiently disseminate information about peers’ topics and relevant peers, we propose new semantic-based gossip protocols. In addition, we propose an efficient query routing algorithm that selects the best peers to recommend documents based on the gossip-view entries and query topics. At the query’s initiator, recommendations are selectively chosen based on similarity, rates and popularity or other recommendation criteria. In our experimental evaluation, using the TREC09 dataset, we show that using semantic gossip increases recall by a factor of 1.6 compared to well-known random gossiping. Furthermore, P2Prec has the ability to get reasonable recall with acceptable query processing load and network traffic.