“Data science and artificial intelligence are going to be determinants of future success in our current global competition,” said NSA Senior Data Science Authority Tony Thrall. “The intelligence community needs to partner even more than ever with both academia and industry.”
Thrall says he was energized by the event’s participation and spirit.
“The NSA wanted to see if we could organize a data science competition for students,” he said Jianwei Niu, UTSA professor of computer science, associate dean of University College and an SDS faculty member. “I leveraged the experience UTSA has had in organizing six or seven hackathons in the past and said, ‘We have a wonderful group of student leaders, so let’s see if they can take on this challenge and organize the first datathon on campus.'”
Students who attended the datathon explored data science questions with their peers and competed to investigate the socioeconomic factors that influence low birth weight and newborn mortality. The challenge was framed as a commission by a fictitious government agency attempting to project these outcomes in Texas in the year 2030.
“The Rowdy Datathon was designed with a perspective not found in other similar events,” he said Juan Gutiérrez, professor and chair of the UTSA Department of Mathematics. “One of the main considerations was to expose data hackers, or ‘dackers’ to the real complexity of data analysis.”
Gutiérrez said there is a common misconception—even among students—that data analysis is just coding. However, he explained that students must learn skills such as data management, including how to cope with large amounts of data, or data with errors, as well as ethics within data management.
The competitors’ projects were judged by Gutiérrez and six data analysts with the NSA. The judges evaluated the teams based on a number of factors such as how well they presented the results, the soundness of their methodologies, and the reproducibility of their results.
“In the end, we want students who participate in this challenge to grow,” Gutiérrez said. “The next iteration will refine areas that proved to need more polishing, and, with this solid foundation, we are well on our way to make the Rowdy Datathon an example to follow and a notable event in the nation.”
Although all participants worked with the same data, they were split into three tracks: beginner, intermediate and advanced. This made the datathon accessible to students of any skill level, explained Ronnie Maddoxa student organizer majoring in environmental science.
“We expect that most of the people who attend these events have either never touched code or they’re freshmen and sophomores and they’ve taken one or two programming classes,” she said. “So even though there is a competitive aspect and we have prizes and awards at the end of the weekend, the emphasis is definitely on learning.”
Although the datathon is designed to be beginner-friendly, more advanced students found plenty to keep them occupied. The event featured workshops and provided students with an invaluable opportunity to network and meet other students and professionals in the field.
Niu believes that accepting students regardless of skill level is especially important in STEM fields such as computer or data science, which she said can often seem daunting or unattainable for many.
“Students often become intimidated when thinking about the field of data science, what data science is and if they should go into it or if they’re able to,” she said. “This datathon definitely offers a great opportunity for students to try. This is the best thing we can do to help enrich the curriculum. It’s outside the classroom and it exposes students to real-world challenges.”
Maddox said the planning team is trying to ensure that as much content as possible from the event will be made accessible to the public in the future. This could include recordings of some of the workshops, or even promoting the dataset and challenges.
The team is also planning for future Rowdy Datathons.
“We still want to create as many opportunities for as much of our community as possible,” Maddox said.
The student organizers said the dedication they have to data science students at UTSA reflects the support the organizers have received from their faculty, advisors and mentors at the university and in the community.
“This isn’t in anybody’s job description,” Millison said, “but it wouldn’t be possible without people who really want to see UTSA data science students succeed and have a big impact.”