Papers by Andres Abeliuk
Detecting Polarized Topics Using Partisanship-aware Contextualized Topic Embeddings (2021.findings-emnlp)
Copied to clipboard
| Challenge: | polarization of the news media has been blamed for fanning disagreement, controversy and even violence. |
| Approach: | They propose a method to automatically detect polarized topics from partisan news sources by corpus-contextualized topic embedding a news corpus on a topic and using cosine distance to capture topical polarization. |
| Outcome: | The proposed method captures topical polarization and shows it can retrieve the most polarized topics. |