Using Online Media Sharing Behavior as Implicit Feedback for Collaborative Filtering

In many practical recommender systems, it is found difficult to obtain explicit feedback from users about the preference for a specific item, such as music, book, movie, etc. Most researches up to this point has focused on tracking various sources of implicit feedback from user behavior including purchase history, browsing patterns, and watching habits, in order to model user preference. In this paper, we investigate a method that uses information exploited from a user's online media sharing activities as a novel source of implicit feedback for recommendation system. We look into elements of media sharing behavior and suggest whether behaviors have the potentiality that could play a role as a predictor of users’ preference. Then in a specific domain, we choose appropriate behaviors by two criteria: abundance and observability. As a representative case, we focus on YouTube, one of the most popular social video sharing sites. By criteria we suggest, we select three behaviors including favorite, upload and view and formulate the simple item-based algorithm based on those behaviors. Through a series of experiments, we evaluate recommendation results obtained from our dataset by comparison with those from other reference algorithms. The results show that favorite and upload have possibility to be used as implicit feedback.

[1]  Shankar Kumar,et al.  Video suggestion and discovery for youtube: taking random walks through the view graph , 2008, WWW.

[2]  John Riedl,et al.  Item-based collaborative filtering recommendation algorithms , 2001, WWW '01.

[3]  Bradley N. Miller,et al.  GroupLens: applying collaborative filtering to Usenet news , 1997, CACM.

[4]  Danah Boyd,et al.  Social Network Sites: Definition, History, and Scholarship , 2007, J. Comput. Mediat. Commun..

[5]  Steve Fox,et al.  Evaluating implicit measures to improve web search , 2005, TOIS.

[6]  Marcelo Maia,et al.  Identifying user behavior in online social networks , 2008, SocialNets '08.

[7]  Guy Shani,et al.  Mining recommendations from the web , 2008, RecSys '08.

[8]  Douglas W. Oard,et al.  Implicit Feedback for Recommender Systems , 1998 .

[9]  David M. Nichols,et al.  Implicit Rating and Filtering , 1998 .

[10]  Tao Luo,et al.  Discovery and Evaluation of Aggregate Usage Profiles for Web Personalization , 2004, Data Mining and Knowledge Discovery.

[11]  David Heckerman,et al.  Empirical Analysis of Predictive Algorithms for Collaborative Filtering , 1998, UAI.

[12]  Mark Claypool,et al.  Implicit interest indicators , 2001, IUI '01.

[13]  Jonathan L. Herlocker,et al.  Evaluating collaborative filtering recommender systems , 2004, TOIS.

[14]  Abigail Sellen,et al.  Music sharing as a computer supported collaborative application , 2001, ECSCW.

[15]  William W. Wadge,et al.  Harnessing the power of "favorites" lists for recommendation systems , 2009, RecSys '09.

[16]  Douglas W. Oard,et al.  Modeling Information Content Using Observable Behavior , 2001 .

[17]  Osmar R. Zaïane,et al.  Combining Usage, Content, and Structure Data to Improve Web Site Recommendation , 2004, EC-Web.

[18]  Susan T. Dumais,et al.  Improving Web Search Ranking by Incorporating User Behavior Information , 2019, SIGIR Forum.

[19]  Yoichi Shinoda,et al.  Information filtering based on user behavior analysis and best match text retrieval , 1994, SIGIR '94.

[20]  Jaime Teevan,et al.  Implicit feedback for inferring user preference: a bibliography , 2003, SIGF.

[21]  Susan T. Dumais,et al.  Learning user interaction models for predicting web search result preferences , 2006, SIGIR.

[22]  Peiling Wang,et al.  A Cognitive Model of Document Use During a Research Project. Study II. Decisions at the Reading and Citing Stages , 1999, Journal of the American Society for Information Science.

[23]  Martin Halvey,et al.  Exploring social dynamics in online media sharing , 2007, WWW '07.