This post is also available in:
עברית (Hebrew)
Disinformation is a growing online phenomenon that is significantly impacting social, political, and economic events and in extreme cases could poses a threat to freedom and democracy. Researchers from IMDEA Networks, Cyprus University of Technology, and LSTECH ESPAÑA SL have developed the HyperGraphDis algorithm that detects disinformation on social media and helps combat the proliferation of fake news.
Author Dr. Marius Paraschiv, Senior Researcher at IMDEA Networks, explained that the study “proposes a detection method that considers the complex social structures between users, as well as relational and semantic elements, to determine the nature of their generated content.”
“With ever-increasing volumes of social media data, achieving high accuracy in detecting fake news is not enough; detection algorithms need to be scalable and fast to handle large volumes of data in near real-time. Our new HyperGraphDis algorithm not only enhances detection accuracy but also significantly reduces execution time, making it much more practical than competing methods,” added Dr. Nikolaos Laoutaris, Research Professor at IMDEA Networks.
According to Techxplore, the scientists evaluated four datasets on the 2016 US presidential elections and the COVID-19 pandemic on X (then Twitter), and found that HyperGraphDis significantly outperformed existing methods in accuracy and computational efficiency.
HyperGraphDis combines advanced techniques like hypergraph neural networks, graph clustering for community detection, and natural language processing for text understanding, all allowing for more efficient and accurate detection of disinformation.
While the researchers focused on X, HyperGraphDis can be adapted to any other social platform. It also offers platform owners an effective way to mitigate the effects of disinformation, providing a better understanding of how it spreads and how to combat it effectively by facilitating fact-verified and contextually tailored responses.
Although the project faced several challenges along the way (including struggling to collect up-to-date data from X due to content being deleted or edited), the researchers are already looking to the future and planning multimodal disinformation detection using advanced models like GPT-4.
