Exploration and Use of Large Model-driven Digital Humans for Popularizing Earthquake Science

Autor: Demeng WU, Shaoxian ZHANG, Hongyu QIAO
Jazyk: English<br />Chinese
Rok vydání: 2024
Předmět:
Zdroj: CT Lilun yu yingyong yanjiu, Vol 33, Iss 5, Pp 655-660 (2024)
Druh dokumentu: article
ISSN: 1004-4140
DOI: 10.15953/j.ctta.2024.079
Popis: Earthquakes in China frequently cause severe damage, underscoring the importance of actively promoting science education to enhance public awareness on earthquake prevention and disaster reduction and proactive disaster mitigation. However, significant gaps in the content, presentation, and richness of current earthquake science-related education exist as well as a lack of public engagement in such efforts, necessitating improvement in all aspects. By leveraging artificial intelligence and large language models, we can develop advanced educational tools to create new innovative methods for enhancing public interactivity and engagement and content richness in earthquake science-related education. The International Seismological Data Center has created QuakeGPT, a large model for the vertical earthquake domain based on Alibaba’s open-source model “Tongyi Qianwen.” This model has acquired extensive professional knowledge related to earthquakes and provides richer, more accurate information through question-and-answer services. QuakeGPT offers earthquake knowledge and disaster prevention advice via the WeChat “XiaoQ” platform, increasing public engagement and learning effectiveness and thereby significantly enhancing the general public’s knowledge on earthquake science and their ability to respond to earthquakes.
Databáze: Directory of Open Access Journals