Constructive Language in News Comments

We discuss the characteristics of constructive news comments, and present methods to identify them. First, we define the notion of constructiveness. Second, we annotate a corpus for constructiveness. Third, we explore whether available argumentation corpora can be useful to identify constructiveness in news comments. Our model trained on argumentation corpora achieves a top accuracy of 72.59% (baseline=49.44%) on our crowd-annotated test data. Finally, we examine the relation between constructiveness and toxicity. In our crowd-annotated data, 21.42% of the non-constructive comments and 17.89% of the constructive comments are toxic, suggesting that non-constructive comments are not much more toxic than constructive comments.

[1]  Joel R. Tetreault,et al.  Abusive Language Detection in Online User Content , 2016, WWW.

[2]  Jürgen Schmidhuber,et al.  Long Short-Term Memory , 1997, Neural Computation.

[3]  Claire Cardie,et al.  A Survey on Assessment and Ranking Methodologies for User-Generated Content on the Web , 2015, ACM Comput. Surv..

[4]  A. Tseronis.  From Connectives to Argumentative Markers: A Quest for Markers of Argumentative Moves and of Related Aspects of Argumentative Discourse , 2011 .

[5]  Douglas Biber,et al.  Variation across speech and writing: Methodology , 1988 .

[6]  Lucas Dixon,et al.  Ex Machina: Personal Attacks Seen at Scale , 2016, WWW.

[7]  Yuzhou Wang,et al.  Locate the Hate: Detecting Tweets against Blacks , 2013, AAAI.

[8]  Brian Ecker,et al.  Argument Mining: Extracting Arguments from Online Dialogue , 2015, SIGDIAL Conference.

[9]  Yue Zhang,et al.  Context-Sensitive Lexicon Features for Neural Sentiment Analysis , 2016, EMNLP.

[10]  Hong Yu,et al.  Bidirectional RNN for Medical Event Detection in Electronic Health Records , 2016, NAACL.

[11]  Anette Frank,et al.  Argumentative texts and clause types , 2016, ArgMining@ACL.

[12]  Dirk Hovy,et al.  Hateful Symbols or Hateful People? Predictive Features for Hate Speech Detection on Twitter , 2016, NAACL.

[13]  Moshe Azar,et al.  Argumentative Text as Rhetorical Structure: An Application of Rhetorical Structure Theory , 1999 .

[14]  Manfred Stede,et al.  Rhetorical structure and argumentation structure in monologue text , 2016, ArgMining@ACL.

[15]  Iryna Gurevych,et al.  Argumentation Mining in User-Generated Web Discourse , 2016, CL.

[16]  Ingmar Weber,et al.  Automated Hate Speech Detection and the Problem of Offensive Language , 2017, ICWSM.

[17]  Ani Nenkova,et al.  Revisiting Readability: A Unified Framework for Predicting Text Quality , 2008, EMNLP.

[18]  Jeffrey Pennington,et al.  GloVe: Global Vectors for Word Representation , 2014, EMNLP.

[19]  Cristian Danescu-Niculescu-Mizil,et al.  Conversational Markers of Constructive Discussions , 2016, NAACL.

[20]  Shafiq R. Joty,et al.  CODRA: A Novel Discriminative Framework for Rhetorical Analysis , 2015, CL.

[21]  Marie-Francine Moens,et al.  Automatic detection of arguments in legal texts , 2007, ICAIL.

[22]  Brink van der Merwe,et al.  Comment classification for an online news domain , 2014 .

[23]  J. Schmidhuber,et al.  Framewise phoneme classification with bidirectional LSTM networks , 2005, Proceedings. 2005 IEEE International Joint Conference on Neural Networks, 2005..

[24]  Joel R. Tetreault,et al.  Finding Good Conversations Online: The Yahoo News Annotated Comments Corpus , 2017, LAW@ACL.

[25]  J. Cornfield,et al.  A method of estimating comparative rates from clinical data; applications to cancer of the lung, breast, and cervix. , 1951, Journal of the National Cancer Institute.

[26]  Frans H. van Eemeren,et al.  Argumentative Indicators in Discourse, A Pragma-Dialectical Study , 2007, Argumentation Library.

[27]  Nicholas Diakopoulos.  Picking the NYT Picks : Editorial Criteria and Automation in the Curation of Online News , 2015 .

[28]  Kalina Bontcheva,et al.  Stance Detection with Bidirectional Conditional Encoding , 2016, EMNLP.