Article by Eric Worrell
“…However, there are still many obstacles to overcome before the goal of automatic unmasking can be achieved…”
Hierarchical machine learning model identifies motivators of climate change misinformation on social media
Communications Earth & Environment Volume 5article number: 436 (2024) Cite this article
abstract
Misinformation about climate change poses a significant threat to society’s well-being, so effective mitigation strategies are urgently needed. However, The rapid proliferation of online misinformation on social media platforms has outstripped the ability of fact-checkers to debunk false claims. Automatic detection of climate change error messages offers a promising solution. In this study, we address this gap by developing a two-step hierarchical model. The Enhanced Computer-Assisted Denial and Doubt (CARDS) model was specifically designed to classify climate claims on Twitter. Additionally, we apply the enhanced card model to 5 million climate-themed tweets over a six-month period in 2022. Peaks in climate contrarianism coincide with one of four stimuli: political events, natural events, contrarian influencers, or convinced influencers. Implications for automatic responses to climate misinformation are discussed.
Learn more: https://www.nature.com/articles/s43247-024-01573-7
The section about the ultimate goal of the project is very interesting.
…These findings have practical implications. adopt Our model can help Twitter/X augment and enhance ongoing manual fact-checking procedures A computer-assisted program is provided to find tweets most likely to contain climate misinformation. This adoption could more effectively detect and counter climate-related misinformation and help Twitter/X enforce policies to reduce false or misleading claims on the platform. However, environmental groups have shown that Twitter/X ranks last among major social media platforms in terms of policies and procedures for combating climate misinformation. and there is little evidence that X will improve these programs in the short term41. Alternatively, our model could provide the basis for an API that Twitter/X users can use to evaluate climate-related claims they see in their feeds. Overall, the potential practical applications of our model highlight the need for continued academic work to monitor misinformation on Twitter/X and raise important questions about the data needed to hold social media platforms accountable for the spread of false claims.
However, There are still many hurdles to overcome before the goal of automatic debunking can be achieved. Effective debunking requires both explaining the relevant facts and exposing the misleading fallacies underlying the misinformation. Contrary climate claims can contain a range of different fallacies, so automatically detecting logical fallacies is another necessary task that, when combined with the CARDS model, can bring us closer to the “holy grail of fact-checking”No. 17. …
Learn more: https://www.nature.com/articles/s43247-024-01573-7
Our old friend John Cook contributed to the paper. According to the paper, he is now a researcher at the rather Orwellian-sounding Melbourne Center for Behavior Change, which appears to be part of the Department of Psychology at the University of Melbourne. The center’s mission statement is “Leverage research and education to create sustainable, lasting changes in behaviour, policy and practice that improve lives, livelihoods and the environment.”.
Back in 2013, John Cook suffered public embarrassment when his collection of gruesome art became public knowledge due to poor hygiene on the site. The artwork features a self-portrait of John Cook in a Nazi uniform and also includes a photo of prominent climate skeptics, including Anthony Watts, dressed as a half-naked Roman gladiator. But I’m sure this unusual artistic bent will in no way interfere with the work Cook does toward the Center for Behavior Change’s mission.”Generate sustainable, lasting behavior change”.
I have no idea who funded this AI paper, but given the author's close ties to major universities in Australia and the UK, I'm pretty confident that tax funding is somewhere in the picture.
Relevant
