Exploring the heterogeneity of social media: dataset and a pilot study

Social media has exploded beyond anyone's wildest imagination, and User Generated Content (UGC) is propagated online at an un-paralleled level. Organizing and associating the heterogeneous UGC data in different social media networks is playing a more and more important role in comprehensive user modeling and social media applications. Since social media data are essentially user-centric, which are generated from user and customized for user, we introduce a user-centric dataset to analyze and exploit the social media heterogeneity, which involves with six popular social media networks and 180,000 overlapped users. A pilot study on the collected dataset is conducted to demonstrate the limitations of traditional content-centric metrics in analyzing the cross-OSN data. Finally, we conclude this work with discussions on the potentials of user-centric framework for cross-OSN organization and association.