How do you expose students to real world problems and solutions? You can bring the real world to them, which is what happened at the recent Datathon held at the U of I campus.

About 250 students came together at the Alice Campbell Alumni Center for a competition coordinated by the Illini Statistics Club, the Department of Statistics, and Synchrony. The students divided up into 75 teams and had 36 hours to predict the stock market activity of several companies for the six weeks of 2019 without using any 2019 data.

Analyzing only news articles and market activity from 2018, the participants used a Natural Language Processing model—a type of artificial intelligence used to read and understand human behavior—along with statistics and computer programs to make their predictions.

All participants were also challenged to present their findings in a creative, visual way. Judges scored teams based on how their predictions compared to what really happened in the stock market and how creative their presentation was. There was also a prize for best data visualization.

“They have to present their findings and come up with a viable solution to the problem as well. Yet, in these 36 hours, students are able to discover rich insights about the dataset and discover high level challenges in the industry,” said Kanika Punhani, president of the Illini Statistics Club.

girl
A student presents at Datathon. About 250 students participated in this year's event. (Photo by Jesse Wallace.)

Punhani also said that participating in Datathon allows students to showcase their industry data and analytical skills to potential employers and build their resume.

“By attending the interactive workshops, students will grasp skills companies want students to have in the workforce. Having the Datathon listed on their resume allows students to showcase to potential employers their experience of working with industry data and analytical skills,” Punhani said.

Yutong Liu, sophomore studying statistics, computer science, and economics, said she learned how important teamwork is by participating in Datathon. Liu was part of the winning best undergraduate team.

“There is no way we can win this challenge without any single one of our teammates. We supported each other and worked on areas we are confident in,” Liu said. She added that it was very interesting to see how differently other teams approached the same problem."

Wanxing Dai, sophomore studying statistics, was also part of Liu’s team. She said she originally wanted to participate in Datathon because she wanted to utilize statistics more in her everyday life. Dai said Datathon taught her to apply her skills outside of the classroom setting and how to work better in a team.

“I learned that when analyzing data, it is important to interpret and demonstrate the information in the appropriate way because each chart tells a different story,” Dai said.

 In addition to the competition, there was also a corporate sponsor meet and greet. Workshops and meals were provided, along with grand prizes that included some of the latest technology gadgets.  

Synchrony, the lead sponsor for the event, worked with the Illini Statistics Club and the Department of Statistics to pioneer Datathon in 2018. This year, three more corporate sponsors joined —Sandia National Laboratories, Bayer, and 3M.

Ujjval Patel, vice president of the emerging technology center at Synchrony, said students have the opportunity to network with a diverse range of industry professionals who can provide a wider lens for new perspectives and capabilities and potential mentors.

Patel said that one of the key components of Datathon was creating a forum for students to learn more data science capabilities. He also spoke about the importance of holding events such as Datathon for students on campus.

“It gives students an opportunity to work outside of textbook examples. Datathon provides real-world examples to help hone and apply their skills,” Patel said.

Punhani said she hopes to expand the event in the future, so that students from all majors can participate in order to challenge themselves and network with professionals in the field.

“Data is never perfect. It’s often messy to the eye; however, the fun lies in revealing patterns and making sense of the numbers,” Punhani said.

Abby Paeth
2019-04-01