Using New York Times Picks to Identify Constructive Comments

Published in Proceedings of NLP meets journalism workshop, 2017

Recommended citation: Varada Kolhatkar and Maite Taboada. 2017. Using New York Times Picks to Identify Constructive Comments. In Proceedings of NLP meets journalism workshop, Association for Computational Linguistics, Copenhagen, Denmark, pages 100-105. https://aclanthology.info/papers/W17-4218/w17-4218

Download paper here

We examine the extent to which we are able to automatically identify constructive online comments. We build several classifiersusing New York Times Picks as positive examples and non-constructive thread comments from the Yahoo News AnnotatedComments Corpus as negative examples of constructive online comments. We evaluate these classifiers on a crowd-annotatedcorpus containing 1,121 comments. Our best classifier achieves a top F1 score of 0.84.