Model Card
Model Details
Model Description
This model identifies 12 distinct climate change denial strategies or fallacies to classify and analyse texts that express skepticism or opposition to climate change scientific findings. These 12 distinct labels come from the FLICC taxonomy created by John Cook and his colleagues. The FLICC taxonomy divides denial strategies into five primary categories: fake experts, logical fallacies, impossible expectations, cherry-picking, and conspiracy theories.
- Developed by: Francisco Zanartu, John Cook, Julian Garcia and Markus Wagner
- Model type: DeBERTa (Decoding-enhanced BERT with disentangled attention) is a Transformer-based neural language model
- Language(s) (NLP): Finetuned and evaluated on a English dataset.
- License: [More Information Needed]
- Finetuned from model: microsoft/deberta-v2-xlarge
Fallacy Type | Definition |
---|---|
Ad hominem | Attacking a person/group instead of addressing their arguments |
Anecdote | Using personal experience or isolated examples instead of sound arguments or compelling evidence |
Cherry Picking | Selecting data that appear to confirm one position while ignoring other data that contradicts that position |
Conspiracy theory | Proposing that a secret plan exists to implement a nefarious scheme such as hiding a truth |
Fake experts | Presenting an unqualified person or institution as a source of credible information. |
False choice | Presenting two options as the only possibilities, when other possibilities exist |
False equivalence | Incorrectly claiming that two things are equivalent, despite the fact that there are notable differences between them |
Impossible expectations | Demanding unrealistic standards of certainty before acting on the science |
Misrepresentation | Misrepresenting a situation or an opponent’s position in such a way as to distort understanding |
Oversimplification | Simplifying a situation in such a way as to distort understanding, leading to erroneous conclusions |
Single cause | Assuming a single cause or reason when there might be multiple causes or reasons |
Slothful induction | Ignoring relevant evidence when coming to a conclusion |
- Downloads last month
- 79
This model does not have enough activity to be deployed to Inference API (serverless) yet. Increase its social
visibility and check back later, or deploy to Inference Endpoints (dedicated)
instead.