Spatiotemporal analysis of anti-Asian hate on social media in the United States
Topics:
Keywords: BERT, hate, Asian, COVID-19
Abstract Type: Paper Abstract
Authors:
Alexander Hohl, The University of Utah
Aggie Yellow Horse, Arizona State University
Richard Medina, The University of Utah
Neng Wan, The University of Utah
Ming Wen, The University of Utah
Xiao Huang, University of Arkansas
Zhenlong Li, University of South Carolina
,
,
,
Abstract
Since the first confirmed case of COVID-19 in the US on January 19, 2020, the anti-Asian racist and xenophobic rhetoric began to surge on social media; followed by acts of discrimination and harassment against Asians and Asian Americans in the US. Our objective is to identify anti-Asian hate language from geotagged social media posts, illustrate their spatiotemporal distribution, and find relationships between socioeconomic and demographic characteristics of places and hate. We identify hateful language in a dataset of 18 million geotagged tweets on the topic of COVID-19 using Bidirectional Encoder Representations from Transformers (BERT), find clusters of hate using LISA statistics, and use spatial regression techniques to establish associations of hate with multiple covariates. Our results can inform decision makers in public health and safety for allocating resources for place-based preparedness and response for the pandemic-induced racism as a public health threat.
Spatiotemporal analysis of anti-Asian hate on social media in the United States
Category
Paper Abstract
