Hands on Learning: Mining the Web for data

On the home page:


Twitter Flu thumbnail

Click on the thumbnail for a larger image.


Aron Culotta, left, talks with students Matthew Gill, center, and Ross Murray about their research project which focuses on tracking disease outbreaks through monitoring of Twitter messsages


Matthew Gill, graduate student, Integrated Science & Technology 

Ross Murray, senior, computer science major, Department of Computer Science & Industrial Technology 

Dr. Aron Culotta, Assistant Professor of Computer Science



Matthew Gill and Ross Murray

Students Ross Murray, left,
and Matthew Gill 

Matthew Gill and Ross Murray are putting their computer science skills to work while learning first hand the art of research collaboration.


Gill, a graduate student from Holden in the Integrated Science and Technology Program, and computer science senior Murray of Slidell, are working with Aron Culotta, assistant professor of computer science, in a research program that monitors the social network program Twitter to determine disease trends such as influenza outbreaks.


The research, supported by a grant from the Louisiana Board of Regents, looked at more than 500 million Twitter messages over an eight-month period. By using Twitter’s application programming interface (API) with a small number of keywords to track rates of influenza-related messages, the team was able to forecast future influenza rates with an amazing 95 percent correlation with national health statistics collected by the Centers for Disease Control.


“A micro-blogging service such as Twitter is a promising new data source for Internet-based surveillance because of the volume of messages, their frequency and public availability,” said Culotta. “This approach is much cheaper and faster than having thousands of hospitals and health care providers fill out forms each week.”


The two students worked closely with Culotta in the collection of the data. They would meet weekly with him to review their work, share observations and tweak any approaches that needed refinement.


Murray had previously worked with his instructor on an unrelated independent study course last summer. He joined the Twitter project when Culotta invited him to work with him.


“Dr. Culotta is great to work with and very accommodating and skilled,” Murray said. “The experience and skills I’ve gained are very valuable for me.”


Gill echoes the sentiment: “Dr. Culotta is simply amazing. He always provides us with new and interesting challenges. I never truly understood what Twitter was or how it worked until now. Working on this team has taught me better programming techniques as well as valuable collaborative skills.”


Their work continues, as Culotta looks to refine his project.


“We have been recreating the previous version of Dr. Culotta’s project in a way that it can be extended to other topics,” Murray said.


“Overall, we’ll rewrite and optimize the program and expand it to incorporate other social media sources,” added Gill.


The two, according to their professor, are excellent workers and collaborate well as a team. “Ross is a star student, a great programmer and thoughtful collaborator,” Culotta said. “Matt is very disciplined and knowledgeable of the field. They are both extremely enthusiastic hard working, easy to collaborate with, and full of great suggestions. I think they enjoy working on an application of computer science that can have a real impact on people.”


More Homepage Stories...