Extreme hazard events can severely threaten the well-being of society.To understand their impacts on the well-being of people,social scientists have proposed various indicators related to individuals’sufering,and ana...Extreme hazard events can severely threaten the well-being of society.To understand their impacts on the well-being of people,social scientists have proposed various indicators related to individuals’sufering,and analyzed them mainly via post-disaster surveys.Social media has shown its value in capturing people’s perceptions of disasters,but few scholars have investigated individuals’sufering levels based on social media data.Accordingly,this study used social media data and developed a hybrid model that combines machine learning classifers and lexicon-based approaches to estimate the wellbeing impacts of disasters,which are measured by individuals’physical,emotional,and social sufering levels.Six machine learning models were trained to categorize the sufering as refected by disaster-related posts on Weibo.Convolutional neural networks(CNN)were found to be the most accurate model,and was selected to classify all posts into four groups(no suffering,physical sufering,emotional sufering,and social sufering).In each classifed group of posts,word co-occurrence analysis was then applied to construct sufering lexicons.By combining the classifcation results from CNN and sufering lexicons,this study proposed optimization-based algorithms to estimate sufering levels for posts across space and time.The proposed model was applied to the 2023 Beijing-Tianjin-Hebei extreme rainfall event.The temporal analysis revealed that individuals’physical sufering levels declined more rapidly than mental and social sufering levels.Spatial analysis revealed that individuals’sufering presented high spatial heterogeneity,and that the hazard-afected regions experienced signifcantly greater levels of sufering.This hybrid model provides an analytical tool for timely and human-centered disaster emergency management.展开更多
基金国家自然科学基金项目(4020100440271007)科技部社会公益项目(2004DIB3J095)"+1 种基金Challenge Programon Water & Food"Conservation agriculture for thedryl-andareas oftheYellow River Basin"国土资源部百名优秀青年科技人才计划等项目资助
基金supported by the National Natural Science Foundation of China(Grant Nos.72304039,52394232)the Fundamental Research Funds for the Central Universities(Grant Nos.2243300007,x2ggC2250200)the National Key R&D Program of China(Grant No.2024YFC3808601).
文摘Extreme hazard events can severely threaten the well-being of society.To understand their impacts on the well-being of people,social scientists have proposed various indicators related to individuals’sufering,and analyzed them mainly via post-disaster surveys.Social media has shown its value in capturing people’s perceptions of disasters,but few scholars have investigated individuals’sufering levels based on social media data.Accordingly,this study used social media data and developed a hybrid model that combines machine learning classifers and lexicon-based approaches to estimate the wellbeing impacts of disasters,which are measured by individuals’physical,emotional,and social sufering levels.Six machine learning models were trained to categorize the sufering as refected by disaster-related posts on Weibo.Convolutional neural networks(CNN)were found to be the most accurate model,and was selected to classify all posts into four groups(no suffering,physical sufering,emotional sufering,and social sufering).In each classifed group of posts,word co-occurrence analysis was then applied to construct sufering lexicons.By combining the classifcation results from CNN and sufering lexicons,this study proposed optimization-based algorithms to estimate sufering levels for posts across space and time.The proposed model was applied to the 2023 Beijing-Tianjin-Hebei extreme rainfall event.The temporal analysis revealed that individuals’physical sufering levels declined more rapidly than mental and social sufering levels.Spatial analysis revealed that individuals’sufering presented high spatial heterogeneity,and that the hazard-afected regions experienced signifcantly greater levels of sufering.This hybrid model provides an analytical tool for timely and human-centered disaster emergency management.
