论文信息 - An Empirical Study on Preference Distribution of Facebook Users

An Empirical Study on Preference Distribution of Facebook Users

Online social networks connect users all around the world,, now is becoming a valuable crowdsourced data source for implementing intelligent people-centric services, e.g., personalized recommendations. Understanding users' characteristics, behaviors is critical for enabling such intelligent services. This paper studies users' preference distribution based on a huge Facebook dataset with 479,048 users, 11 preference types, 879,868 distinct preference items, 6,276,422 preference items. We observe that Music, TV, Movie are the most popular preference types on Facebook, users share more common items in TV, Movie, Sport, Music than the other preference types. We also find that the preference popularity distribution follows power law. In addition, we investigate the effects of users' location on the distribution of the popular preferences, conclude that the most popular preferences are very different among different countries. For instance, only one same Music ("Michael Jackson"), one same TV ("Family Guy"), no same Movie present as the top 10 preference items in all of US, Denmark, Germany.

Xiao Han | Leye Wang | Longbiao Chen

[1] Ryohei Orihara,et al. Metrics for Evaluating the Serendipity of Recommendation Lists , 2007, JSAI.

[2] Gediminas Adomavicius,et al. Toward the next generation of recommender systems: a survey of the state-of-the-art and possible extensions , 2005, IEEE Transactions on Knowledge and Data Engineering.

[3] Dan Cosley,et al. Do social explanations work?: studying and modeling the effects of social explanations in recommender systems , 2013, WWW.

[4] Gayatri Swamynathan,et al. Do social networks improve e-commerce?: a study on social marketplaces , 2008, WOSN '08.

[5] Mao Ye,et al. Location recommendation for location-based social networks , 2010, GIS '10.

[6] Niklas Carlsson,et al. Evolution of an online social aggregation network: an empirical study , 2009, IMC '09.

[7] Daqing Zhang,et al. The Emergence of Social and Community Intelligence , 2011, Computer.

[8] Sean M. McNee,et al. Improving recommendation lists through topic diversification , 2005, WWW '05.

[9] Chen-Nee Chuah,et al. Unveiling facebook: a measurement study of social network based applications , 2008, IMC '08.