Twitter Inc (NYSE:TWTR) will be used for a study to help researchers to understand human emotions. Researchers in Australia have plans to create real-time online emotional maps of entire cities and countries. This will be done through the mining of millions of Tweets on Twitter.
‘We Feel’ – the project
The Commonwealth Scientific and Industrial Research Organisation (CSIRO) has created and unveiled software that has the capacity of going through 500 tweets per second and analyzing the emotional tone that will amount to around 27 million Tweets per day covering numerous Twitter Inc (NYSE:TWTR) accounts. Emotions such as joy, love, anger, surprise, fear and sadness all have some specific set of words associated with them and the software searches through its database of 600 words that are connected with these emotions. It also tries to determine the location by tracking the time zones connected with the tweets from various Twitter accounts.
The project has been named ‘We Feel’, and according to Dr Cecile Paris, who is a social media analytics expert at CSIRO, the project will make available a huge amount of information on the emotional state of entire regions in real-time that is useful for mental health researchers.
“It could uncover who is at risk, what our emotions depend on, whether we are different in different regions, how we fluctuate over time,” the expert added. Time-intensive surveys with small sample sizes form the basis of current research methods and the project will help in improving those too.
Software could help Twitter users
Other software designed to assess large scale emotions has been created by the Black Dog Institute in tandem with Amazon Web Services, but still has to go through the validation process. ‘Black Dog Index’ is a survey conducted for evaluating the emotional well being of the Australian population that is ongoing, and the results of ‘We Feel’ project will be compared with this index for validation.
Black Dog Institute’s executive director Helen Christensen says that this software will be helpful in quickly supplying the information that otherwise takes several years for collection.
“I don’t think anyone yet fully understands how powerful a tool social media can be in the mental health field,” she said.
The software results could also be helpful in detecting behavior that reflects a depressed individual. Despair among people or an individual could potentially be detected. For Twitter Inc (NYSE:TWTR) users, if the language used in tweets indicates that the user could be suicidal, then certain automated responses could be generated by the software.