Data Visualization for Supporting Linguists in the Analysis of Toxic Messages

The goal of this research is to provide linguists with visualisations for analysing the results of their hate speechannotation. These visualisations consist of a set of interactive graphs for analysing the global distribution ofannotated messages, finding relationships between features, and detecting inconsistencies in the annotation.We used a corpus that includes 1,262 comments posted in response to different Spanish online new articles.The comments were annotated with features such as sarcasm, mockery, insult, improper language, construc-tivity and argumentation, as well as with level of toxicity (’not-toxic’, ’mildly toxic’, ’toxic’ or ’very toxic’).We evaluated the selected visualisations with users to assess the graphs’ comprehensibility, interpretabilityand attractiveness. One of the lessons learned from the study is the usefulness of mixed visualisations that in-clude simple graphs (Bar, Heat map) - to facilitate the familiarisation with the results of the annotated corpustogether with more complex ones (Sankey, Spider or Chord) - to explore and identify relationships betweenfeatures and to find inconsistencies.