Big Data Is Only What We Make of It
In the final installment of the College of Science lecture series, the UA's Vincent J. Del Casino Jr. explains what defines big data, what its future holds and what it means for higher education.
What do you think you know about big data?
Many might say the phrase describes countless numbers that can be analyzed to show patterns or trends about humans and society.
But it might not be that simple.
In a lecture Monday night at Centennial Hall, Vincent J. Del Casino Jr. suggested big data may be only what we make of it. In reality, he said, the term describes minuscule, abstract pieces of information — "small moments in time," he called them — that only make sense when we use it as a lens to view society.
"Big data is only made real by us," said Del Casino, the UA's vice president of Academic Initiatives and Student Success and a professor in the School of Geography and Development. "Big data becomes a thing by our relationship to it and the questions asked of it."
Del Casino's talk was the last of six in a weekly public lecture series presented by the College of Science called "Humans, Data and Machines." The series, the 14th in the last 13 years, tackled the subject of artificial intelligence and the convergence of the digital, physical and biological worlds. To close out the series, Del Casino explained what big data really is, what its future looks like, how it will impact society, and how higher education can leverage it to revolutionize teaching and learning.
Del Casino kicked off the lecture by defining "big data" — keyword "big" — as a massive collection of "bits and bytes" that moves at a high rate of speed through the internet and machines. Think about the 87,000-plus hours of video streamed on Netflix every minute. And thanks to the internet and cellphones, that data is being collected in real time and then quickly individualized, helping companies such as Amazon deliver up-to-the-minute suggestions on which book you should buy next.
This form of unstructured data is far different from the structured kind that scientists collect and store in databases to use for research, Del Casino said.
"That variety is actually producing opportunity, but also producing all kinds of challenges, I think, to us as humans," he added.
Chief among those challenges, Del Casino said, is big data's tendency to pollute the information environment. To illustrate this, Del Casino pointed to new facial-recognition software that can help create manufactured videos of people saying or doing things they didn't really say or do.
The final third of Del Casino's lecture explored the implications that big data has for higher education. Machine learning has increased personalization in higher education much in the same way it has for other industries, which has led to universities and colleges having to prove their value to society. As a result, public support and funding are declining, Del Casino said.
"We're in this kind of complex spiral where we're going to be put under pressure to produce this kind of personalized learning, but we're not necessarily going to have the funds to do it," he said.
Del Casino said data and machine learning can be leveraged to support education, creating things such as open educational resources that the UA and other institutions are using. These programs that use technology and machine learning to deliver education to people who might not otherwise have access could help solidify higher education's value to the public in the future, Del Casino said.
"There's really cool opportunities to leverage these technologies to do some of the things that aren't always easy at a distance," he said.
Del Casino said higher education as an industry should continue leveraging technology to create learning communities focused on improving literacy in a variety of areas, such as technology, data, creativity and inclusiveness. Organizing education around these literacy skills will prepare students for a variety of possible careers, as opposed to the current system that asks students to choose a major for a narrow career path within that discipline.
"We have to challenge the notion that one major leads to one career," he said. "If we organize around literacy skills, majors lead to all kinds of careers, and careers lead to all kinds of lifelong learning. We need to organize ourselves in reaction to this sort of world in order to really effect change and bring people back so they're continuing to move themselves along."
- Jan. 22: "Problem Solving With Algorithms," Stephen Kobourov, UA professor of computer science
- Jan. 29: "The Minds of Machines," Mihai Surdeanu, UA associate professor of computer science
- Feb. 5: "Working Alongside Thinking Machines," Nirav Merchant, director, UA Data Science Institute
- Feb. 12: "What Humans Do That Machines Cannot," Luis von Ahn, CEO and co-founder, Duolingo; professor of computer science, Carnegie Mellon University
- Feb. 19: "Machine Influencers and Decision Makers," Jane Bambauer, UA professor of law
University of Arizona in the News