Twitter Used To Detect Changes In Public Mood
Brett Smith for Redorbit.com
One of the unintended consequences of Twitter is the ability to quantify conversations and statements, albeit 140 characters at a time. This allows researchers and analysts the opportunity to comb mountains of Twitter data and parse wisdom and theories from their findings.
A new UK research study by academics at the University of Bristol‘s Intelligent Systems Laboratory have linked public mood via Twitter postings to events of social upheaval and change based on analyzed data from the micro-blogging site.
Researchers said they observed a significant increase in negative mood with the announcement of spending cuts and during summer riots last year. Conversely, they found an increase in positive postings surrounding the time of last year’s royal wedding ceremony between Kate Middleton and Prince William.
A primary point of analysis was the data surrounding the time the British government announced massive spending cuts as a way of coping with the international recession. The study showed not only that a significant mood change on Twitter can be measured, but public mood in general has still not recovered from that announcement.
This could be useful in determining if the public has psychologically overcome traumatic events or if it is still basking in the afterglow of an event filled with civic pride.
The study also points to another important time period in the UK. When riots broke out in various cities around the country in the summer 2011, researchers methods seems to suggest that some increase in public anger foreshadowed these events. Although the significance of this finding is still a matter of dispute, it suggests that Twitter could serve as a “canary-in-the-coal-mine” for social disturbances and unrest.
Another goal of the study was to see if social events translated into mood trends that could be measured by looking at the frequency of words that are correlated with different feelings. In addition to looking at unique social events, researchers also they observed that periodic events such as Christmas, Valentine’s Day and Halloween stir up similar Twitter sentiments, year after year.
In the study, 484 million tweets generated by more than 9.8 million users from the UK were analyzed between July 2009 and January 2012. Researchers said their future work will include the comparison of social media content with traditional media content, as well as the comparisons with traditional opinion polls.
This was the latest in a series of studies that attempted to utilize data culled from Twitter to predict or analyze trends. The studies often look at Twitter as the reflections of an ever-thinking “hive mind.”
In March, a University of California at Riverside professor looked to Twitter to help predict the traded volume and value of a financial stock for the following day. Using a trading model based on statistical data culled from the micro-blogging site, the professor and a small team of researchers were able to outperform the Dow Jones Industrial Average, according to their report. Alos, a study at the USC Annenberg Innovation Lab inaccurately attempted to predict this year’s Academy Award winners after a year of collecting data from Twitter.
