On Understanding the Divergence of Online Social Group Discussion

We study online social group dynamics based on how group members diverge in their online discussions. Previous studies mostly focused on the link structure to characterize social group dynamics, whereas the group behavior of content generation in discussions is not well understood. Particularly, we use Jensen-Shannon (JS) divergence to measure the divergence of topics in user-generated contents, and how it progresses over time. We study Twitter messages (tweets) in multiple real-world events (natural disasters and social activism) with different times and demographics. We also model structural and user features with guidance from two socio-psychological theories, social cohesion and social identity, to learn their implications on group discussion divergence. Those features show significant correlation with group discussion divergence. By leveraging them we are able to construct a classifier to predict the future increase or decrease in group discussion divergence, which achieves an area under the curve (AUC) of 0.84 and an F-1 score (harmonic mean of precision and recall) of 0.8. Our approach allows to systematically study collective diverging group behavior independent of group formation design. It can help to prioritize whom to engage with in communities for specific topics of needs during disaster response coordination, and for specific concerns and advocacy in the brand management.

[1]  H. Tajfel,et al.  Social categorization and intergroup behaviour , 1971 .

[2]  Daniel M. Romero,et al.  Influence and passivity in social media , 2010, ECML/PKDD.

[3]  Daniel L. Wann,et al.  The Positive Social and Self Concept Consequences of Sports Team Identification , 1991 .

[4]  Ana-Maria Popescu,et al.  A Machine Learning Approach to Twitter User Classification , 2011, ICWSM.

[5]  Ani Nenkova,et al.  Automatically Assessing Machine Summary Content Without a Gold Standard , 2013, CL.

[6]  D. R. White,et al.  Structural cohesion and embeddedness: A hierarchical concept of social groups , 2003 .

[7]  Henriette Cramer,et al.  Performing a check-in: emerging practices, norms and 'conflicts' in location-sharing using foursquare , 2011, Mobile HCI.

[8]  F. Harary,et al.  The cohesiveness of blocks in social networks: Node connectivity and conditional density , 2001 .

[9]  Srinivasan Parthasarathy,et al.  Scalable graph clustering using stochastic flows: applications to community discovery , 2009, KDD.

[10]  Jun Zhu,et al.  User grouping behavior in online forums , 2009, KDD.

[11]  Robert E. Kraut,et al.  Increasing commitment to online communities by designing for social presence , 2011, CSCW.

[12]  John D. Lafferty,et al.  Dynamic topic models , 2006, ICML.

[13]  F. Reid.  Rediscovering the Social Group: A Self-Categorization Theory , 1987 .

[14]  Filippo Menczer,et al.  The Geospatial Characteristics of a Social Movement Communication Network , 2013, PloS one.

[15]  David Yarowsky,et al.  Hierarchical Bayesian Models for Latent Attribute Detection in Social Media , 2011, ICWSM.

[16]  D. Beal,et al.  Cohesion and performance in groups: a meta-analytic clarification of construct relations. , 2003, The Journal of applied psychology.

[17]  B. Tuckman.  DEVELOPMENTAL SEQUENCE IN SMALL GROUPS. , 1965, Psychological bulletin.

[18]  Jon M. Kleinberg,et al.  Group formation in large social networks: membership, growth, and evolution , 2006, KDD '06.

[19]  Jure Leskovec,et al.  The life and death of online groups: predicting group growth and longevity , 2012, WSDM '12.

[20]  B. Mullen,et al.  The relation between group cohesiveness and performance: An integration. , 1994 .

[21]  Jianfeng Gao,et al.  An Information-Theoretic Approach to Automatic Evaluation of Summaries , 2006, NAACL.

[22]  T. Kamalanabhan,et al.  Unearthed: The Other Side of Group Cohesiveness , 2005 .

[23]  A. Lott,et al.  Group cohesiveness as interpersonal attraction: a review of relationships with antecedent and consequent variables. , 1965, Psychological bulletin.

[24]  Víctor M. Eguíluz,et al.  Distinguishing topical and social groups based on common identity and bond theory , 2013, WSDM.

[25]  Jianhua Lin,et al.  Divergence measures based on the Shannon entropy , 1991, IEEE Trans. Inf. Theory.

[26]  S. Kiesler,et al.  Applying Common Identity and Bond Theory to Design of Online Communities , 2007 .

[27]  Rakesh Agrawal,et al.  On participation in group chats on Twitter , 2013, WWW.

[28]  Omar Alonso,et al.  User Taglines: Alternative Presentations of Expertise and Interest in Social Media , 2012, 2012 International Conference on Social Informatics.

[29]  Amit P. Sheth,et al.  Prediction of Topic Volume on Twitter , 2012 .