{"id":425792,"date":"2020-04-20T16:44:49","date_gmt":"2020-04-20T20:44:49","guid":{"rendered":"http:\/\/www.rochester.edu\/newscenter\/?p=425792"},"modified":"2020-04-20T16:50:31","modified_gmt":"2020-04-20T20:50:31","slug":"can-twitter-predict-attacks-against-asians-425792","status":"publish","type":"post","link":"https:\/\/www.rochester.edu\/newscenter\/can-twitter-predict-attacks-against-asians-425792\/","title":{"rendered":"Can Twitter anticipate attacks against Asians and Asian Americans?"},"content":{"rendered":"

University of Rochester computer scientists are gleaning a wealth of information from Twitter users to document the social impacts of the novel coronavirus pandemic.<\/p>\n

For example, a new study<\/a> by the research group of Jiebo Luo, a professor of computer science, and posted to the scholarly website ArXiv<\/a>, finds that the increased use of terms like \u201cChinese virus\u201d and \u201cWuhan virus\u201d on the social media platform\u00a0<\/span>correlated strongly with a rise in media reports of attacks on Chinese and other Asians.<\/p>\n

The researchers were also able predict with more than 80 percent accuracy which Twitter users are more likely to use the terms based on their age, gender, geographic location, \u201csocial capital,\u201d and political affiliation. The terms used to refer to the source of the pandemic has sparked controversy in some media circles between those who consider a geographic description an accurate reflection of where the virus originated while others consider the geographic terms to be pejorative.<\/p>\n

A real-time look at a large-scale crisis <\/strong><\/h3>\n
\"A
A timeline shows the density of global online media coverage using controversial terms (in red) and global online media coverage of COVID-19-related racial attacks (in blue). For the most part, the rises and dips correspond.<\/figcaption><\/figure>\n

\u201cTo the best of our knowledge, this is the first large-scale social media-based study to characterize users with respect to their usage of controversial terms during a major crisis,\u201d writes lead author Hanjia Lyu, a PhD student. Long Chen \u201920, an undergraduate in the group, is a co-author along with Luo.<\/p>\n

Luo\u2019s group is also using Twitter data to explore other aspects of the coronavirus pandemic, including its impact on mental health, on the success of crowd-funding platforms, on how college students react to social distancing, and the relationship between hoarding and scarcity.<\/p>\n

\u201cThe data captured in social media platforms can provide an important real-time look into how people communicate and what they think is important to talk about,\u201d Luo says.\u00a0<\/span><\/p>\n

The researchers gathered more than 17 million tweets\u2014about 1.5 terabytes of data\u2014from March 23 to 26. They then applied a facial recognition platform to help determine which Twitter users could be confidently characterized by age, gender, and race. Users who followed candidates from both parties were excluded.\u00a0<\/span><\/p>\n

This produced a working database of 593,233 tweets using \u201ccontroversial terms\u201d and 490,168 tweets using \u201cnoncontroversial terms.\u201d\u00a0<\/span><\/p>\n

The researchers then used machine-learning classifier techniques to predict which users would be most likely to use either controversial or noncontroversial terms.\u00a0<\/span><\/p>\n

 <\/p>\n

\"A
Based on analysis of tweets from 1,083,401 Twitter users, Luo\u2019s lab correlated these characteristics with likelihood users would use controversial or noncontroversial terms. Blue bars extending farthest above 0 indicate highest likelihood of using controversial terms. Blue bars extending farthest below 0 indicate highest likelihood of using noncontroversial terms.<\/figcaption><\/figure>\n

Suburban as well as rural users most likely to use controversial terms<\/strong><\/h3>\n

The researchers were able to draw a number of conclusions based on their analysis of the more than one million tweets. Among them:<\/p>\n