Computer science and statistics students sat huddled together, heads close together, animatedly debating which company they wanted to work for.
BYU’s big-data capstone course partners students with real-world clients and provides them with hands-on work experience with massive sets of data. The class’s kickoff event was held at Vivint headquarters in Lehi, Utah, and students were introduced to this year’s clients and the projects they prepared for students.
“It’s . . . real-world problems that I don’t think you get a ton of in traditional classes,” Taylor Redd said. “You have to piece together everything that you’ve learned to do.”
Redd is an adjunct professor, who makes up one-third of the capstone’s professorial team. Redd works as a data science consultant at Adobe while mentoring capstone students part-time. Redd teaches the statistics side of the class, while professors Giraud-Carrier and Snell focus on computer science.
“[The class] is two sections, one that is predominantly stats students and one that is predominantly CS students,” Redd explained. “But we’re all kind of one big class. We do everything together.”
Students are organized into teams of four, usually with both statistics and computer science majors in the mix. Then, based on the presentations from clients, each team lists which projects they would most like to work on for the rest of the year.
Some of this year’s clients include Lawrence Livermore National Labs, National Instruments, Vivint, Qualtrics, FamilySearch, and U of U Health Sciences.
“It’s a lot of really cool stuff,” computer science student Dallin Cox said. “It seems like a lot of the projects are pretty big.”
Students involved in the BYU Store’s project would use sales/stock data from its physical and online shopping forums to build a prediction model to determine the amount of stock stores should hold. According to the BYU Store’s director, the capstone group’s work could save the organization nearly two million dollars each year.
Students could also have the opportunity to work with NUVI—a social media analysis company that analyzes over 100 million social media posts daily. Capstone participants have been asked to work with this enormous amount of data and develop a process to identify which social media posts contain feedback regarding an organization’s products or services.
Some other projects include working with geodata to improve Vivint’s smart home-assistant app, improving Ziff’s facial-recognition software, and developing machine-learning algorithms to identify types of questions written by Qualtrics users.
“I’ve done a couple internships, and the one thing that I really liked about those is the fact that when I did a project . . . it was actually providing some meaningful service,” Cox said. “I felt like taking this class . . . [because] I felt like it would be a bit more realistic and practical.”
After each group selected their top-five preferences, they were assigned to work on a project until the end of winter semester.