Sign In

Communications of the ACM

ACM TechNews

Twitter Firehose Reveals How Weather Affects Mood

View as: Print Mobile App Share:
Correlating tweets with weather in different geographies.

Researchers at Stanford University are mining geotagged tweets for indications of mood and correlations with the weather.

Credit: Technology Review

Stanford University researchers have mined geotagged tweets for indications of mood and searched for correlations with the weather.

They studied a database of tweets geotagged to one of 32 major urban areas in the United States, which had been filtered from a dataset of 10 percent of all tweets posted in 2010 and 2011. The researchers also filtered out tweets about major national or international events, and then categorized the remaining tweets as either anger-hostility, fatigue-inertia, depression-dejection, or sleepiness-freshness. Finally, they used a machine-learning algorithm to find correlations with the weather in these areas using a database from the U.S. National Oceanic and Atmospheric Administration, which records the average daily temperature, the daily change in temperature, the daily rainfall or snowfall, the depth of snow, and the total amount and strength of the sun each day.

"People tend to be happier as temperature becomes cooler but feel uncomfortable with drastic temperature decrease," notes Stanford researcher Jiwei Li.

However, the researchers caution these correlations are not necessarily an indication of causation. For example, the data shows snow correlates with negative moods, but it is unclear whether this is a result of the weather itself or other factors caused by the weather, such as an increase in traffic jams or accidents.

From Technology Review
View Full Article


Abstracts Copyright © 2014 Information Inc., Bethesda, Maryland, USA


No entries found