Exploiting Diversification in Gossip-Based Recommendation - Data Management in Cloud, Grid and P2P Systems - page 25

Exploiting Diversification
in Gossip-Based Recommendation
Maximilien Servajean 1 , Esther Pacitti 1 , Miguel Liroz-Gistau 2 ,
Sihem Amer-Yahia 3 , and Amr El Abbadi 4
1 INRIA & LIRMM, University of Montpellier, France
{ servajean,pacitti } @lirmm.fr
2 INRIA & LIRMM, Montpellier, France
miguel.liroz [e-mail]
3 CNRS, LIG
[e-mail]
4 Dpt. of Computer Science, University of California at Santa Barbara
[e-mail]
Abstract. In the context of Web 2 . 0, the users become massive pro-
ducers of diverse data that can be stored in a large variety of systems.
The fact that the users' data spaces are distributed in many different
systems makes data sharing di cult. In this context of large scale dis-
tribution of users and data, a general solution to data sharing is offered
by distributed search and recommendation. In particular, gossip-based
approaches provide scalability, dynamicity, autonomy and decentralized
control. Generally, in gossip-based search and recommendation, each user
constructs a cluster of “relevant” users that will be employed in the pro-
cessing of queries. However, considering only relevance introduces a sig-
nificant amount of redundancy among users. As a result, when a query
is submitted, as the user profiles in each user's cluster are quite simi-
lar, the probability of retrieving the same set of relevant items increases,
and recall results are limited. In this paper, we propose a gossip-based
search and recommendation approach that is based on a new clustering
score, called usefulness , that combines relevance and diversity, and we
present the corresponding new gossip-based clustering algorithm. We val-
idate our proposal with an experimental evaluation using three datasets
based on MovieLens , Flickr and LastFM . Compared with state of the
art solutions, we obtain major gains with a three order of magnitude
recall improvement when using the notion of usefulness regardless of the
relevance score between two users used.
1 Introduction
In the context of Web 2 . 0, users become massive producers of diverse data ( e.g.
photos, videos, scientific data) that can be stored in a large variety of systems
( e.g. DropBox, Facebook, Flickr, Google+, local computer or smartphone). Users
Work conducted within the Institut de Biologie Computationnelle and partially
funded by the labex NUMEV and the CNRS project Mastodons.