TY - GEN
T1 - FakeSens
T2 - 17th Annual International Conference on Distributed Computing in Sensor Systems, DCOS 2021
AU - Kou, Ziyi
AU - Shang, Lanyu
AU - Zhang, Yang
AU - Youn, Christina
AU - Wang, Dong
N1 - Funding Information:
This research is supported in part by the National Science Foundation under Grant No. IIS-2008228, CNS-1845639, CNS-1831669, Army Research Office under Grant W911NF-17-1-0409. The views and conclusions contained in this document are those of the authors and should not be interpreted as representing the official policies, either expressed or implied, of the Army Research Office or the U.S. Government. The U.S. Government is authorized to reproduce and distribute reprints for Government purposes notwithstanding any copyright notation here on.
Publisher Copyright:
© 2021 IEEE.
PY - 2021
Y1 - 2021
N2 - Social sensing is emerging as an effective and pervasive sensing paradigm to collect timely data and observations from human sensors. This paper focuses on the problem of COVID-19 misinformation detection on social media. Our work is motivated by the lack of COVID-specific knowledge in current misinformation detection solutions, which is critical to assess the truthfulness of social media claims about the emerging COVID-19 disease. In this paper, we leverage human intelligence on a crowdsourcing platform to obtain essential knowledge facts for detecting the COVID-19 misinformation on social media. Two critical challenges exist in solving our problem: i) how to efficiently acquire accurate and timely knowledge that is both inclusive and specific to COVID-19? ii) How to effectively coordinate the efforts from both expert and non-expert workers to detect COVID-19 misinformation? To address these challenges, we develop FakeSens, a social sensing based crowd knowledge graph approach that explicitly explores the knowledge facts specific to COVID-19 and models the reliability of different types of crowd workers to capture the misleading COVID-19 claims. Evaluation results on a real-world dataset show that FakeSens significantly outperforms state-of-the-art baselines in accurately detecting misleading claims of COVID-19 on social media.
AB - Social sensing is emerging as an effective and pervasive sensing paradigm to collect timely data and observations from human sensors. This paper focuses on the problem of COVID-19 misinformation detection on social media. Our work is motivated by the lack of COVID-specific knowledge in current misinformation detection solutions, which is critical to assess the truthfulness of social media claims about the emerging COVID-19 disease. In this paper, we leverage human intelligence on a crowdsourcing platform to obtain essential knowledge facts for detecting the COVID-19 misinformation on social media. Two critical challenges exist in solving our problem: i) how to efficiently acquire accurate and timely knowledge that is both inclusive and specific to COVID-19? ii) How to effectively coordinate the efforts from both expert and non-expert workers to detect COVID-19 misinformation? To address these challenges, we develop FakeSens, a social sensing based crowd knowledge graph approach that explicitly explores the knowledge facts specific to COVID-19 and models the reliability of different types of crowd workers to capture the misleading COVID-19 claims. Evaluation results on a real-world dataset show that FakeSens significantly outperforms state-of-the-art baselines in accurately detecting misleading claims of COVID-19 on social media.
UR - http://www.scopus.com/inward/record.url?scp=85123283263&partnerID=8YFLogxK
UR - http://www.scopus.com/inward/citedby.url?scp=85123283263&partnerID=8YFLogxK
U2 - 10.1109/DCOSS52077.2021.00035
DO - 10.1109/DCOSS52077.2021.00035
M3 - Conference contribution
AN - SCOPUS:85123283263
T3 - Proceedings - 17th Annual International Conference on Distributed Computing in Sensor Systems, DCOS 2021
SP - 140
EP - 147
BT - Proceedings - 17th Annual International Conference on Distributed Computing in Sensor Systems, DCOS 2021
PB - Institute of Electrical and Electronics Engineers Inc.
Y2 - 14 July 2021 through 16 July 2021
ER -