Misinformation Detection Methods¶
Computational approaches and technical strategies for identifying misinformation, disinformation, and fake news. Methods range from shallow linguistic features (n-grams, lexical properties) to deep neural architectures (RNNs, CNNs, graph neural networks) to structural analysis (discourse trees, propagation patterns).
Key papers¶
- Learning Hierarchical Discourse-level Structure for Fake News Detection — discourse structure learning via unsupervised dependency trees; 82.19% accuracy
- EANN: Event Adversarial Neural Networks for Multi-Modal Fake News Detection — multimodal neural detection
- Detect Rumors in Microblog Posts Using Propagation Structure via Kernel Learning — propagation-based detection with tree kernels