Annotating scientific uncertainty: A comprehensive model using linguistic patterns and comparison with existing approaches

Panggih Kusuma Ningrum, Philipp Mayr, Nina Smirnova, Iana Atanassova

Abstract: Highlights•We propose a method to identify and categorise scientific uncertainty in scholarly full text.•Our system UnScientify, using rule-based and linguistic processing, achieves accuracy of 0.8.•UnScientify outperforms Large Language Models and Deep learning models.•The rule-based approach provides interpretable results.