综合信息疫情监测系统：以韩国的新冠疫情为例。

Integrated Infodemic Surveillance System: The Case of COVID-19 in South Korea.

作者信息

Park Gil-Sung, Bae Jintae, Lee Jong Hun, Yun Byung Yeon, Lee Byunghwee, Shin Eun Kyong

机构信息

Department of Sociology, Korea University, South Korea.

Department of Physics, Korea Advanced Institute of Science and Technology, South Korea.

出版信息

Stud Health Technol Inform. 2021 May 27;281:1036-1040. doi: 10.3233/SHTI210342.

DOI:10.3233/SHTI210342

PMID:34042836

Abstract

This study merges multiple COVID-19 data sources from news articles and social media to propose an integrated infodemic surveillance system (IISS) that implements infodemiology for a well-tailored epidemic management policy. IISS is an à-la-carte infodemic surveillance solution that enables users to gauge the epidemic related consensus, which compiles epidemic-related data from multiple sources and equipped with various methodological toolkits - topic modeling, Word2Vec, and social network analysis. IISS can provide reliable empirical evidence for proper policymaking. We demonstrate the heuristic utilities of IISS using empirical data from the first wave of COVID-19 in South Korea. Measuring discourse congruence allows us to gauge the distance between the discourse corpus from different sources, which can highlight consensus and conflicts in epidemic discourse. Furthermore, IISS detects discrepancies between social concerns and main actors.

摘要

本研究整合了来自新闻文章和社交媒体的多个新冠疫情数据源，提出了一个综合信息疫情监测系统（IISS），该系统运用信息流行病学来制定精准的疫情管理政策。IISS是一种定制化的信息疫情监测解决方案，能让用户评估与疫情相关的共识，它整合了来自多个来源的疫情相关数据，并配备了各种方法工具包——主题建模、词向量模型和社交网络分析。IISS可为合理的政策制定提供可靠的实证依据。我们利用韩国第一波新冠疫情的实证数据展示了IISS的启发式效用。衡量话语一致性使我们能够评估不同来源的话语语料库之间的差异，这可以突出疫情话语中的共识和冲突。此外，IISS还能检测社会关注与主要行为者之间的差异。