Tweet 'sentiment analysis' predicts Hillary Clinton win

Scientists, including one of Indian-origin, have developed a new software which can automatically determine how someone feels based on what they write or say on Twitter, and predicts that Hillary Clinton will win the US presidential election.
Researchers from the University of Utah in the US used the software - sentiment analysis - to analyse the individual sentiments of more than 1.6 million geo-tagged tweets about the US presidential election over the last five months.
A database of these tweets was then examined to determine whether states and their counties are leaning towards the Republicans or Democrats.
"With sentiment analysis, it will try to predict the emotions behind every human being when he or she is talking or writing something," said project leader Debjyoti Paul, doctoral student at the University of Utah.

"With that in mind, we are not just trying to look at the information in the tweets. We are trying to incorporate the emotion with the information," said Paul.
The team has created an interactive website in which users can find out if the tweets coming out of their state and its counties are more positive or negative toward Republicans or Democrats during any defined period of time since June 5.

Also, the data can tell the percentage of both positive and negative tweets towards a political party and when there was a surge for a particular type of tweet in the last five months.

Based on the number of positive tweets posted since June towards each party, the computer model predicts that Democratic Party nominee Clinton will win the US presidential election.
It also suggests that Republicans sent out 17 per cent more political tweets than Democrats.
Delaware was the only state in which a majority of tweets from all counties in the state were positive towards the same party - in this case, the Democrats, researchers said.
For the Republicans, South Dakota had the highest percentage of counties in which most of their tweets were positive towards the party (73 per cent of the counties).

"Not only did the number of positive tweets for Democrats peak after the last two debates and the Trump federal taxes story, it is also when the most negative tweets about the Democratic Party were posted," researchers said.
Paul and his team started with more than 250 million tweets posted around the world from June 5 to October 30 this year and then weeded out all non-political tweets based on a system of keywords using advanced software.
Then those tweets were sifted through the software where each tweet was analysed and assigned a score from 0 to 1 where 0 is the most negative sentiment, 1 is the most positive sentiment, and 0.5 is neutral.