A paper coauthored by PhD student Lanyu Shang and members of Associate Professor Dong Wang’s research group, the Social Sensing and Intelligence Lab within the School of Information Sciences on the University of Illinois Urbana-Champaign, received the very best paper award within the research track throughout the 2022 IEEE/ACM International Conference on Advances in Social Network Evaluation and Mining (ASONAM 2022). The conference, which was held in Istanbul, Turkey, on November 10-13, brings together researchers and practitioners from a broad number of social media-related fields to advertise collaborations and exchange of ideas and practices.

Their paper, “A Knowledge-driven Domain Adaptive Approach to Early Misinformation Detection in an Emergent Health Domain on Social Media,” addresses a vital problem of how one can accurately detect misinformation in emergent health domains, where existing misinformation detection solutions often fall short in training effective classification models as a consequence of the dearth of sufficient training data and up-to-date medical knowledge.

“In our study, we observe that misinformation from the emergent health domain of Monkeypox is usually relevant to topics in recent news, akin to COVID-19. For instance, a well-liked misleading post within the Monkeypox domain claims that the Monkeypox virus is intentionally engineered for the financial interest of vaccine sale like COVID-19,” the researchers noted. “While such misinformation can’t be easily detected solely with our previous knowledge in regards to the Monkeypox disease, our COVID-19 knowledge, akin to the COVID-19 virus just isn’t engineered, will be of great help for debunking the above Monkeypox misinformation.”

Inspired by the above remark, the researchers explored the wealthy and timely resources, akin to the annotated data and medical reports, from a relevant health domain, namely COVID-19, to tackle the early misinformation detection problem within the emergent health domain of Monkeypox. Based on the researchers, through experiments on multiple real-world datasets, the proposed framework was shown to be effective in identifying emergent healthcare misinformation in an early stage.

The authors consider that their work is also applied to detect health misinformation related to other emergent health domains, akin to polio, respiratory syncytial virus (RSV), or any diseases in the long run. The accurate and timely misinformation detection results can effectively mitigate the spread of online misinformation related to emerging diseases and make sure the credibility of knowledge on social media.

The first research focus of the Social Sensing and Intelligence Lab lies within the emerging area of human-centered AI, AI for social good, and cyber-physical systems in social spaces. The lab develops interdisciplinary theories, techniques, and tools for fundamentally understanding, modeling, and evaluating human-centered computing and knowledge (HCCI) systems, and for accurately reconstructing the proper “state of the world,” each physical and social.

Source:

University of Illinois School of Information Sciences