In one of the world’s largest hubs for pollution, data on air
toxins is relatively slim. In China, tracking pollution’s implications in all
but bigger cities can be impossible because local governments oftentimes refuse
to release data to the public.
Researchers at the University of
Wisconsin-Madison have attempted to address the problem with an innovative
solution. Without the data to track, the team has followed complaints related
to air quality on social media.
"There's not enough information about pollution, and
sometimes people suffer from heavier air pollution. We wondered, 'How can we
use a new information source to help people understand [the severity of] the
pollution around?'" said graduate student Shike Mei in a CBS interview,
who, along with Han Li and Jing Fan Mei, conducted a study published in the
IEEE/ACM International Conference on Advances in Social Network Analysis and
Mining in August.
The team tapped into the Twitter-like site Sina Weibo for
posts related to air quality. They developed a machine-learning model that
would recognize posts that contained terms that might suggest poor air quality
on any given day. Words like “haze,” “indoors” and “pollution” indicated bad
air, while terms like “sunshine” indicated clearer conditions.
The model uses those word choice indicators and the location of
their authors to estimate the air quality above a given city or region.
No comments:
Post a Comment