Web 2.0 applications have attracted a considerable amount of attention because their open-ended nature allows users to create light-weight semantic scaffolding to organize and share content. To date, the interplay of the social and semantic components of social media has been only partially explored. Here we focus on Flickr and Last.fm, two social media systems in which we can relate the tagging activity of the users with an explicit representation of their social network. We show that a substantial level of local lexical and topical alignment is observable among users who lie close to each other in the social network. We introduce a null model that preserves user activity while removing local correlations, allowing us to disentangle the actual local alignment between users from statistical effects due to the assortative mixing of user activity and centrality in the social network. This analysis suggests that users with similar topical interests are more likely to be friends, and therefore semantic similarity measures among users based solely on their annotation metadata should be predictive of social links. We test this hypothesis on the Last.fm data set, confirming that the social network constructed from semantic similarity captures actual friendship more accurately than Last.fm's suggestions based on listening patterns.
Deep Dive into Folks in Folksonomies: Social Link Prediction from Shared Metadata.
Web 2.0 applications have attracted a considerable amount of attention because their open-ended nature allows users to create light-weight semantic scaffolding to organize and share content. To date, the interplay of the social and semantic components of social media has been only partially explored. Here we focus on Flickr and Last.fm, two social media systems in which we can relate the tagging activity of the users with an explicit representation of their social network. We show that a substantial level of local lexical and topical alignment is observable among users who lie close to each other in the social network. We introduce a null model that preserves user activity while removing local correlations, allowing us to disentangle the actual local alignment between users from statistical effects due to the assortative mixing of user activity and centrality in the social network. This analysis suggests that users with similar topical interests are more likely to be friends, and there
Folks in Folksonomies:
Social Link Prediction from Shared Metadata
Rossano Schifanella1∗
Alain Barrat2,3
Ciro Cattuto3
Benjamin Markines4
Filippo Menczer3,4
1 Department of Computer Science, University of Turin, Italy
2 Centre de Physique Théorique (CNRS UMR 6207), Marseille, France
3 Complex Networks and Systems Laboratory, ISI Foundation, Turin, Italy
4 School of Informatics and Computing, Indiana University, Bloomington, IN, USA
ABSTRACT
Web 2.0 applications have attracted a considerable amount of at-
tention because their open-ended nature allows users to create light-
weight semantic scaffolding to organize and share content. To date,
the interplay of the social and semantic components of social me-
dia has been only partially explored. Here we focus on Flickr and
Last.fm, two social media systems in which we can relate the tag-
ging activity of the users with an explicit representation of their
social network. We show that a substantial level of local lexical
and topical alignment is observable among users who lie close to
each other in the social network. We introduce a null model that
preserves user activity while removing local correlations, allowing
us to disentangle the actual local alignment between users from sta-
tistical effects due to the assortative mixing of user activity and cen-
trality in the social network. This analysis suggests that users with
similar topical interests are more likely to be friends, and therefore
semantic similarity measures among users based solely on their an-
notation metadata should be predictive of social links. We test this
hypothesis on the Last.fm data set, confirming that the social net-
work constructed from semantic similarity captures actual friend-
ship more accurately than Last.fm’s suggestions based on listening
patterns.
Categories and Subject Descriptors
H.1.2 [Information Systems]: Models and Principles—Human informa-
tion processing; H.3.5 [Information Storage and Retrieval]: Online Infor-
mation Services—Web-based services; H.5.3 [Information Interfaces and
Presentation]: Group and Organization Interfaces—Collaborative comput-
ing, Web-based interaction
General Terms
Algorithms, Experimentation, Measurement
∗Corresponding author. Email: schifane@di.unito.it. A good por-
tion of the work in this paper was carried out while Dr. Schifanella was a
visiting scholar at the Center for Complex Networks and Systems Research
(CNetS) of the Indiana University School of Informatics and Computing.
Permission to make digital or hard copies of all or part of this work for
personal or classroom use is granted without fee provided that copies are
not made or distributed for profit or commercial advantage and that copies
bear this notice and the full citation on the first page. To copy otherwise, to
republish, to post on servers or to redistribute to lists, requires prior specific
permission and/or a fee.
WSDM’10, February 4–6, 2010, New York City, New York, USA.
Copyright 2010 ACM 978-1-60558-889-6/10/02 ...$10.00.
Keywords
Web 2.0, social media, folksonomies, collaborative tagging, social
network, lexical and topical alignment, link prediction, social se-
mantic similarity, Maximum Information Path
1.
INTRODUCTION
Social networking systems like Facebook and systems for con-
tent organization and sharing such as Flickr and Delicious have
created information-rich ecosystems where the cognitive, behav-
ioral and social aspects of a user community are entangled with
the underlying technological platform. This opens up new ways to
monitor and investigate a variety of processes involving the inter-
action of users with one another, as well as the interaction of users
with the information they process. Social media supporting tag-
ging [14, 3] are especially interesting in this respect because they
stimulate users to provide light-weight semantic annotations in the
form of freely chosen terms. Usage patterns of tags can be em-
ployed to monitor interest, track user attention, and investigate the
co-evolution of social and semantic networks.
While the emergence of conventions and shared conceptualiza-
tions has attracted considerable interest [24, 16, 25, 2], little atten-
tion has been devoted so far to relating, at the microscopic level, the
usage of shared tags with the social links existing between users.
The present paper aims at filling this gap. To this end we focus on
Flickr and Last.fm, as to our knowledge they are currently the only
popular social media system where: (1) a significant fraction of the
users provide tag metadata for their content (photographs or songs),
and (2) an explicit representation of the social links between users
is readily available.
The main question that we address in this study is the follow-
ing: given two randomly chosen users, how does the alignment of
their tag vocabularies relate to their proximity on the social net-
work? That is, does lexical alignment exist between neighboring
users, and if so, how does this alignment fade when cons
…(Full text truncated)…
This content is AI-processed based on ArXiv data.