Two UCF alumni and a student each earned first place in a competition that stretched their statistical math skills to come up with better resources to help people visiting theme parks avoid long lines.
Kanak Choudhury ‘17,Taha Mokfi ’17 and statistical computing major Phuong Pho competed individually in Touring Plans’ Big Data Challenge. The contest challenged college students to build statistical models that would predict the wait times for multiple Walt Disney World attractions up to a year in advance. The rides included Splash Mountain, Soarin’ and others. They each will receive from $500 to $1,500 depending on the success of the final models submitted.
The students had to use Big Data to do solve the problem. The information included thousands of files containing wait times for each of the attractions over several years, opening and closing times, temperature records and more.
Pho, a graduate teaching assistant in the economics department, said the project was one of the most time-consuming he’s worked on. He said he tried numerous statistical models until he found one that worked.
“I really enjoyed the Big Data Challenge because it gave me a great opportunity to apply the modelling methodology I learned in school to real-life problems,” he said. “It’s satisfying when you find the meaningful function form or the interaction that improved the predictability of the model.”
Choudhury, who will pursue a doctoral degree in statistics at Iowa State University this fall, said: “It was a really challenging task to find the best model due to large and complicated data. Strong statistical experience helped me to apply the necessary techniques to decipher some of the numbers.”
Mokfi, a statistical computing graduate, used predictive models to accomplish the objective. He said he combined five different models.
“Without having enough theoretical background, no one can be successful in such competitions,” Mokfi said. “UCF offers various courses about data mining and machine learning and these courses can be excellent resources for learning big data methodologies.”
Mokfi is now employed in Hartford, Connecticut, as a data science analyst for Aetna insurance company.
The competition is another example of the power big data yields in helping business and in this case families, plan for the future. A family with young children visiting a theme park for example could use predictive information to plan their day by either avoiding peak waiting times for the children’s favorite rides or by preparing by making sure to have lunch before a long wait.
But big data is still an evolving field and the companies providing these services are looking for talent, especially people who can come up with more efficient and creative approaches.
Touring Plans is a national company that provides park information – including wait times – for tourist attractions via an app. Touring Plans and the UCF Department of Statistics sponsored the competition. The three UCF winners said they are scheduled to meet with Touring Plans officials to discuss the possibility of incorporating one or more of the models on the company’s app.