Track papers across labs, topics, and trends.
Search papers, labs, and topics across Lattice.
This paper introduces a neurosymbolic model for propaganda detection that combines fastText embeddings with symbolic features representing genre, topic, and persuasion techniques. The goal is to improve robustness and generalization compared to LM-based approaches that often overfit training data biases. Experiments demonstrate that incorporating these symbolic features enhances classification performance and robustness compared to text-only methods.
Propaganda detection gets a boost from genre, topic, and persuasion features, outperforming text-only models and suggesting a path to more robust news classification.
Among news disorders, propagandist news are particularly insidious, because they tend to mix oriented messages with factual reports intended to look like reliable news. To detect propaganda, extant approaches based on Language Models such as BERT are promising but often overfit their training datasets, due to biases in data collection. To enhance classification robustness and improve generalization to new sources, we propose a neurosymbolic approach combining non-contextual text embeddings (fastText) with symbolic conceptual features such as genre, topic, and persuasion techniques. Results show improvements over equivalent text-only methods, and ablation studies as well as explainability analyses confirm the benefits of the added features. Keywords: Information disorder, Fake news, Propaganda, Classification, Topic modeling, Hybrid method, Neurosymbolic model, Ablation, Robustness