Sheffield-Trento System for Sentiment and Argument Structure Enhanced Comment-to-Article Linking in the Online News Domain
Ahmet Aker, Fabio Celli, Adam Funk, Emina Kurtić, Mark Hepple, Robert Gaizauskas · 2015
In this paper we describe and evaluate an approach to linking readers ’ comments to online news articles. For each com-ment that is linked based on its comment, we also determine whether the commenter agrees, disagrees or stays neutral with re-spect to what is stated in the article, as well as what the commenter’s sentiment towards the article is. We use similarity features to link comments to relevant arti-cle segments and Support Vector Regres-sion models for assigning argument struc-ture and sentiment. Our results are com-pared to competing systems that took part in MultiLing OnForumS 2015 shared task, where we achieved best linking scores for English and second best for Italian. 1