This post originally appeared on EngagingCities.
It seems that cities known for their successful public transportation programs have done a bit of listening to residents to make improvements, plan, and respond to citizen needs. In a city like Chicago where each day thousands of people rely on the Chicago Transit Authority to get to their destination, a good transit experience seems to make or break their day. And how do we know this? Search for the Twitter hashtag #cta and you’ll discover a multitude of statements from the Chicagoland populace.
Last year, three researchers from Purdue University in Indiana decided to find out if bus and train agencies could learn more about their services by looking closely at Twitter. Craig Collins, Samiul Hasan, and Satish Ukkusuri tested the idea on tweeters who ride Chicago’s rapid transit service, known as the ‘L.’
After all, those who are deeply unsatisfied with something in their city may not always take the time to fill out a survey or be part of a focus group. Why not answer questions like “Do our riders find their commute efficient? Do they truly feel safe riding the ‘L?’ in another way: Each day, Twitter users have their smart phones in hand during commutes -- providing fellow riders with tips concerning a delay or venting about an experience.
The researchers decided to look at publicly available time-stamped Twitter data including geographic location tagging for tweets from those riding the ‘L’ and talking about their experience. They used tweets over seven days last July, weeded out extraneous data, and analyzed the content against a sentiment algorithm with ratings for 298 positive posts and 465 negative ones, on a scale from -5 to +5.
Graphs are posted here for July 4, a holiday and heavy-travel day in particular because of holiday events downtown. The positive tweets are represented by blue lines, the negative by red lines. The researchers also constructed a word cloud from tweets on that day. As you can guess from the trending words, a fire caused huge delays on the Blue line. In the morning, trees fell down on the track delaying the Brown, Purple, and Red lines.
Hasan, a Ph.D. candidate at Purdue, recently presented these findings at the annual Transportation Research Board conference in Washington, dedicated to promoting innovation in transportation. Some were surprised at the lack of positive comments from riders.
“The most interesting thing we found is that transit riders do not give any positive sentiment at a particular time. They only give negative sentiment,” he said. This might not sound like a good thing for the CTA.” But that’s not very disappointing,” Hasan revealed, “because we found that the lack of negative sentiment is basically what transit authorities should look for. If there’s no negative sentiment at any given time, that means that things are running smoothly.”
The CTA receives e-mails and tweets from riders (the CTA began utilizing Twitter last November) but they could also use this type of analysis to cull comments on Twitter to respond to complaints and bolster long term planning efforts. Word clouds in particular could help the CTA see the bigger picture quickly and react to it in real time. Just imagine the knowledge planning or transportation agencies could gain from keeping an eye on public sentiment, and subsequently enhance their infrastructure and services with citizens in mind.
Featured image courtesy of Flickr user clarkmaxwell.
It seems that cities known for their successful public transportation programs have done a bit of listening to residents to make improvements, plan, and respond to citizen needs. In a city like Chicago where each day thousands of people rely on the Chicago Transit Authority to get to their destination, a good transit experience seems to make or break their day. And how do we know this? Search for the Twitter hashtag #cta and you’ll discover a multitude of statements from the Chicagoland populace.
Last year, three researchers from Purdue University in Indiana decided to find out if bus and train agencies could learn more about their services by looking closely at Twitter. Craig Collins, Samiul Hasan, and Satish Ukkusuri tested the idea on tweeters who ride Chicago’s rapid transit service, known as the ‘L.’
After all, those who are deeply unsatisfied with something in their city may not always take the time to fill out a survey or be part of a focus group. Why not answer questions like “Do our riders find their commute efficient? Do they truly feel safe riding the ‘L?’ in another way: Each day, Twitter users have their smart phones in hand during commutes -- providing fellow riders with tips concerning a delay or venting about an experience.
The researchers decided to look at publicly available time-stamped Twitter data including geographic location tagging for tweets from those riding the ‘L’ and talking about their experience. They used tweets over seven days last July, weeded out extraneous data, and analyzed the content against a sentiment algorithm with ratings for 298 positive posts and 465 negative ones, on a scale from -5 to +5.
Graphs are posted here for July 4, a holiday and heavy-travel day in particular because of holiday events downtown. The positive tweets are represented by blue lines, the negative by red lines. The researchers also constructed a word cloud from tweets on that day. As you can guess from the trending words, a fire caused huge delays on the Blue line. In the morning, trees fell down on the track delaying the Brown, Purple, and Red lines.
Hasan, a Ph.D. candidate at Purdue, recently presented these findings at the annual Transportation Research Board conference in Washington, dedicated to promoting innovation in transportation. Some were surprised at the lack of positive comments from riders.
“The most interesting thing we found is that transit riders do not give any positive sentiment at a particular time. They only give negative sentiment,” he said. This might not sound like a good thing for the CTA.” But that’s not very disappointing,” Hasan revealed, “because we found that the lack of negative sentiment is basically what transit authorities should look for. If there’s no negative sentiment at any given time, that means that things are running smoothly.”
The CTA receives e-mails and tweets from riders (the CTA began utilizing Twitter last November) but they could also use this type of analysis to cull comments on Twitter to respond to complaints and bolster long term planning efforts. Word clouds in particular could help the CTA see the bigger picture quickly and react to it in real time. Just imagine the knowledge planning or transportation agencies could gain from keeping an eye on public sentiment, and subsequently enhance their infrastructure and services with citizens in mind.
Featured image courtesy of Flickr user clarkmaxwell.
Org references
Story Category