Workshop on Search in Social Media 2010

Birds of a tag flock together

Our goal: to help the user navigate in a social tagging graph • Navigation-based query construction Serge Abiteboul, Sihem Amer-Yahia, Alban Galland, Amélie Marian, Pierre Senellart • Online incremental query execution • Offline and online ranking and clustering User cloud

+ @serge + #social - #marketing in

#ACMPapers #SpringerPapers

Tag cloud Item cloud AmerYahia09Building

Chi07Understanding Harvey09Folksonomic SantosNeto07Tracking

Marlow06HT06

Oldenburg08Similarity

Joshi09Improving Li07Towards

User partition Tag partition Item partition

Another query example: query(@u,(top-10) $X) = Friends(@serge,@u) Near(G,$X,@u) Near(G,$X,#CarlaBruni) Near(G,$X,#Sarkozy)

CiteULike, delicious, MovieLens

Social tagging data R(user,tag,item)

#WSDM

#SSM

#SIGMOD

Top-k ranking

Affinity between users, tags and items

Social graph analysis

Cluster selection Clusters of users, tags and items

Labeled clustering

Online Offline

• Labeled clustering: adaptation of subspace clustering to social graph • Affinity: domain-dependent definition based on • Cluster selection: efficient evaluation based on experiments on real dataset user interaction • Top-k: incremental query evaluation based on user • User navigation: affinity-based ranking and interaction zooming Challenges and on-going work