Data mining of tweets tracks our mood cycles

CareerMash on TwitterIt’s no surprise Cornell researchers found we’re happiest on weekends but the fact mood cycles were gathered from 500 million tweets shows our casual online chatter is gold to analysts mining data.

Cornell researchers devised a data mining software program to analyze emotions from positive and negative words within English tweets from 2.4 million people in 84 countries. Besides being happiest on weekends, they found good moods run out of steam mid-day and again late in the evening. That wasn’t just for week days before the stress of work brings people down, and before we go to bed, but also on weekends, leading researchers to tie mood changes to our biological clock and sleep patterns.

Although the findings aren’t revolutionary on their own, they did bring a new credibility to the increasing ways that social media data is being analyzed to pick trends in everything from political trends to market research on what music or movies we like.

“It illustrates a new opportunity for doing social and behavioural science in ways that were really unimaginable even five years ago,” says Cornell sociologist Michael Macy in a Globe and Mail story.

Sound like an interesting career? Check out the Data Analytics Specialist career on CareerMash.

Add new comment