This article is interesting from the perspective of critical literacies, as it's based on the idea of recognizing different types of constructive and non-constructive online comments. And it's also interesting from a pedagogical perspective, as it provides a framework that could be used to guide people toward how to make more constructive contents (indeed, having a cohort go through such a labeling exercise would be effective, in my view). Finally, it's interesting because we can imagine the corpus of 12,000 annotated news comments could be used to inform or train automated comment moderation systems.
Today: 0 Total: 7 [Direct link] [Share]
