A new Twitter study by a group of researchers at the University of Ottawa could help City Hall better service the community.
Professor Abdulmotaleb El Saddik, the lead researcher of the study, said he hopes comparing emotions in tweets to their geographical locations could help people contribute to discussions in City Hall. He said he believes it would allow more people to participate directly or indirectly in the decision-making process.
“For example, if there is a discussion about building a new bridge between Ottawa and Gatineau, people can tweet ‘yes’ or ‘no,’ and those sentiments can go directly as input to the counsel of the city,” he said.
El Saddik said his research could also be used if there were traffic complaints on Twitter. If the emotions and locations of the tweets were monitored, possible adjustments could be made to help solve road conditions or traffic lights, he said.
“There is no obligation to take part, but at least someone at City Hall would be listening and knowing the feelings of the people,” said El Saddik.
His study analyzed more than 132,700 tweets from Dec. 4 to Dec. 10. The research team used machines to look for hashtags that may express feelings, such as #blessed, #happy or #sad. They also looked at emoticons within tweets.
Their study compared the tweets of six Canadian cities: Ottawa, Toronto, Vancouver, Edmonton, Halifax, and Montreal.
The same study gained recent media attention from the CBC and Global News, where El Saddik had suggested that Edmonton is the “least happy” Canadian city on Twitter.
Carleton University students say they have difficulty seeing how this research would be controlled and monitored for City Hall.
“I don’t think the execution of the idea would be very intelligent because (with Twitter) you would be getting everybody in the city with their own opinion. It would work best if they front a couple of major opinions and have people vote on those,” said Kent Pederson, a second-year political science student.
“Twitter is an immediate social entity, and people’s responses can be influenced by a number of different things . . . It would be an interesting poll, but I don’t think it should be anything official,” he added.
El Saddik said he understands his study is not yet complete.
“One week of data is not significant enough, at least not to represent the whole year. We have to analyze more data to make a stronger statement,” he said. “We are collecting data at this time, to validate the result we have and to see which event might trigger different results.”
More data about the same cities is planned to be released by El Saddik and his team at the end of January.
