ANTI-SOCIAL COMPUTING:

                                    Explaining, predicting and preventing anti-social behavior (trolling, personal attacks, etc.)                                 

                                                    

KEY PUBLICATIONS   

                                    Conversations Gone Awry: Detecting Early Signs of Conversational Failure

                                    Justine Zhang, Jonathan P. Chang, Cristian Danescu-Niculescu-Mizil,

                                    Lucas Dixon, Yiqing Hua, Nithum Thain, Dario Taraborelli

                                    Proceedings of ACL 2018.


                                    Characterizing Online Public Discussions Through Patterns of Participant Interactions

                                    Justine Zhang, Cristian Danescu-Niculescu-Mizil, Christy Sauper, Sean Taylor

                                    Proceedings of CSCW 2018.


                                    Anyone Can Become a Troll: Causes of Trolling Behavior in Online Discussions

                                    Justin Cheng, Michael Bernstein, Cristian Danescu-Niculescu-Mizil, Jure Leskovec

                                    Proceedings of CSCW, 2017.

                                    Best Paper Award.

                                   

                                    Antisocial Behavior in Online Discussion Communities

                                    Justin Cheng, Cristian Danescu-Niculescu-Mizil, Jure Leskovec

                                    Proceedings of ICWSM, 2015.

                                    Honorable Mention


                                    How Community Feedback Shapes User Behavior

                                    Justin Cheng, Cristian Danescu-Niculescu-Mizil, Jure Leskovec

                                    Proceedings of ICWSM, 2014.


                                    Linguistic Harbingers of Betrayal:  A Case Study on an Online Strategy Game

                                    Vlad Niculae, Srijan Kumar, Jordan Boyd-Graber, Cristian Danescu-Niculescu-Mizil

                                    Proceedings of ACL, 2015.


                                    A computational approach to politeness with application to social factors

                                    Cristian Danescu-Niculescu-Mizil, Moritz Sudhof, Dan Jurafsky, Jure Leskovec, Christopher Potts

                                    Proceedings of ACL, 2013.

                                    Nominated for the Best Paper Award



RELEVANT CODE

                                    ConvoKit: toolkit for analyzing conversations (gone awry)


                                    Code for extracting simple linguistic features from conversations


                                    Stanford Politeness API


                                    Politeness Web App



RELEVANT DATA

                                    Dataset of conversations gone awry


                                    Stanford Politeness Corpus


                                    Interactions in online Diplomacy games


                                                                       

MEDIA COVERAGE     

                                    Interviewed by BBC on Behaving Better Online  (write-up version here)


                                    Our work on causes of trolling was covered by The Wall Street Journal, The Atlantic and The Times      


                                    Our work on antisocial behavior was featured on BBC


                                    Our work on antisocial behavior was covered by The Daily Dot, The Guardian, Wired and others


                                    Our work on betrayal was featured in the The Wall Street Journal                                   

                                   

                                    Our work on betrayal was covered by Science News, CNN, New York Magazine and others


                                    Data Mining Reveals How The “Down-Vote” Leads To A Vicious Circle Of Negative Feedback