Title: Context in Informational Bias Detection
Speaker: Esther Van den Berg (ICL)
Abstract
Informational bias is bias conveyed through sentences or clauses that provide tangential, speculative or background information that can sway readers' opinions towards entities. By nature, informational bias is context-dependent, but previous work on informational bias detection has not explored the role of context beyond the sentence. We present experiments with four kinds of context for informational bias in English news articles: neighboring sentences, the
full article, articles on the same event from other news publishers, and articles from the same domain (but potentially different events). We find that integrating event context improves classification performance over a very strong baseline. In addition, we perform the first error analysis of models on this task. We find that the est-performing context-inclusive model outperforms the baseline on longer sentences, and sentences from politically centrist articles.
