ERICA WONG/The hoya Williams College professor of statistics Richard de Veaux answered questions about the future of big data in Fisher Colloquium last Thursday.
Williams College professor of statistics Richard de Veaux answered questions about the future of big data in Fisher Colloquium last Thursday.

Big data is trending. From company’s customers to their new hires, big data has become a benchmark analytical tool for large businesses looking to get the most bang for their buck. But is it here to stay?
Williams College professor of statistics Richard De Veaux answered this question during an event sponsored by the Undergraduate Business Program Office titled “Big Data — Fool’s Gold or the Mother Lode?”, held in the McDonough School of Business’s Fisher Colloquium last Thursday.
De Veaux holds degrees from Princeton and Stanford and has consulted for a number of Fortune 500 companies including Hewlett-Packard, Pillsbury and General Electric during his 25-year career. He proposed that big data, known as the application of analytics to large data sets in order to find trends in consumer behavior, is something that is increasingly exciting and relevant. “Statisticians are now the cool kids on campus,” he said. “We are sexy and cool. It has finally happened.”
Big data is loosely defined as information “so big that it is a pain,” according to De Veaux. Companies, statisticians and data scientists sort and analyze these massive compilations in order to predict consumer behavior. Data mining allows the analyzing of such large data sets.
Big data is not a new idea, dating back to 1985, although it is a recent buzz term. De Veaux first encountered the concept while consulting for the now defunct First USA Bank.
At the time, banks were collecting large amounts of data to better analyze consumers. “They didn’t collect it to learn secrets of the universe, they collected it for transactions,” De Veaux said. By sorting customers and looking at patterns in consumer behavior, USA Bank attempted to “find out what consumers wanted before they knew what they wanted,” according to De Veaux.
During the 1980s, and still today, big data remains immensely difficult to sort and to then use to reach insightful conclusions.
But an increase in the amount of data collected does not necessarily translate into more accurate insights. Data can easily be interpreted to show false or misleading information. The challenge arises not in collecting and storing data, but analyzing the compiled sets at rapid speeds.
“[Data] doesn’t mean anything unless you can communicate it to someone else, and this can be a challenge,” De Veaux said.
Big data is no different than any other data set except in that it can be exponentially larger.
“I don’t expect to get the answers to what’s going on from this kind of analysis,” De Veaux said. “You need to have standard tools. Big data can give you clues. Big data has big potential, but following it blindly, big data equals big mistakes.”
Freddy Rosas (COL ’15) realized just how transformative big data could be after attending the presentation.
“The days of data growth are nowhere near their end” he said. “Dick De Veaux definitely proved to me, and I’m sure to many others, that much important work lies ahead in the ever-accumulating analysis of big data.”
Jill Weakland (MSB ’15) was impressed by De Veaux’s ability to make big data a more comprehensible topic.
“I loved how De Veaux presented both the good and bad sides of Big Data and made it accessible to everyone, regardless of their previous knowledge,” she said. “He made me want to take a statistics class!”

Have a reaction to this article? Write a letter to the editor.

Leave a Reply

Your email address will not be published. Required fields are marked *