Serendipity-Based Recommendation Framework for SNS Users Using Tie Strength and Relation Clustering

As contents are overflowing in Social Network Services (SNSs), the Recommender System (RS) in SNS became increasingly important. Traditional RSs focus on the relevance of contents to the users and therefore recommend obvious contents over and over again. To solve this problem, many researches have sought to find serendipity, but they have the limitation of recommending obvious or absurd posts. In this paper, we propose a novel method to recommend serendipity using tie strength of the users’ social relationships. Through the implementation of this method, serendipity can be recommended without analyzing user preferences or contents. We developed an illustrative example to prove validity of our framework.

[1]  Eric Gilbert,et al.  Predicting tie strength with social media , 2009, CHI.

[2]  Christos Faloutsos,et al.  TANGENT: a novel, 'Surprise me', recommendation algorithm , 2009, KDD.

[3]  Eli Pariser,et al.  The Filter Bubble: What the Internet Is Hiding from You , 2011 .

[4]  Sean M. McNee,et al.  Being accurate is not enough: how accuracy metrics have hurt recommender systems , 2006, CHI Extended Abstracts.

[5]  Shuaiqiang Wang,et al.  A survey of serendipity in recommender systems , 2016, Knowl. Based Syst..

[6]  Eli Pariser.  FILTER BUBBLE: Wie wir im Internet entmündigt werden , 2012 .

[7]  Nicholas Jing Yuan,et al.  Who Will Reply to/Retweet This Tweet?: The Dynamics of Intimacy from Online Social Interactions , 2016, WSDM.

[8]  Gayle S. Stever,et al.  Twitter as a way for celebrities to communicate with fans: Implications for the study of parasocial interaction. , 2013 .

[9]  Diyi Yang,et al.  Serendipitous Personalized Ranking for Top-N Recommendation , 2012, 2012 IEEE/WIC/ACM International Conferences on Web Intelligence and Intelligent Agent Technology.

[10]  Pasquale Lops,et al.  Introducing Serendipity in a Content-Based Recommender System , 2008, 2008 Eighth International Conference on Hybrid Intelligent Systems.

[11]  Arun Sundararajan,et al.  Trust and the Strength of Ties in Online Social Networks: An Exploratory Field Experiment , 2017, MIS Q..

[12]  Jennifer Marmo,et al.  The rules of Facebook friendship , 2012 .

[13]  Kuei-Hong Lin,et al.  A Social Network-based serendipity recommender system , 2011, 2011 International Symposium on Intelligent Signal Processing and Communications Systems (ISPACS).

[14]  Christopher J. Fariss,et al.  Inferring Tie Strength from Online Directed Behavior , 2013, PloS one.

[15]  Arindam Banerjee,et al.  Semi-supervised Clustering by Seeding , 2002, ICML.

[16]  Doo-Kwon Baik,et al.  Personalized recommender system based on friendship strength in social network services , 2017, Expert Syst. Appl..

[17]  Xiangjian He,et al.  User relationship strength modeling for friend recommendation on Instagram , 2017, Neurocomputing.

[18]  Takayuki Akiyama,et al.  Proposal and Evaluation of Serendipitous Recommendation Method Using General Unexpectedness , 2010, PRSAT@RecSys.

[19]  Wei Dong,et al.  Mutual information: inferring tie strength and proximity in bipartite social network data with non-metric associations , 2012 .

[20]  Huifang Ma,et al.  Combining tag correlation and user social relation for microblog recommendation , 2017, Inf. Sci..

[21]  Sahin Albayrak,et al.  User-centric evaluation of a K-furthest neighbor collaborative filtering recommender algorithm , 2013, CSCW.