ne beautiful autumn
morning in 1986, Goutam
Chakraborty stepped outside
a classroom at the University
of Iowa. The business grad student men-
tally sorted through his own personal big
data ﬁle that had propelled him into a
stark, new life.
Months earlier, the young marketing
executive had been chaueured around
his hometown of Calcutta, India. His
employer, a tobacco company, also paid
for his luxury apartment and handed
him a $10,000 travel and entertainment
“What the heck have I done to myself?
What have I done?” he said to himself
At 27, he had eectively tossed those
amenities out the polished limo door to
become another poor college student
eating ramen noodles in a $250/month
“I thought I’d made the biggest mistake of
my life,” the Oklahoma State University
marketing professor says with a laugh,
sitting in his neat oce in the Business
Building on the Stillwater campus.
The founder of the data mining certiﬁ-
cate program in OSU’s Spears School of
Business says he had moved to Iowa City
to stay just long enough to get a mas-
ter’s degree in statistics and a doctorate
in marketing. Then he’d return home
But, unexpectedly, his passion kicked in
to help students make sense of compli-
cated data and use it to help businesses
and possibly even the world. And that
“It was unplanned. … Most of my fellow
schoolmates back in engineering school
still don’t believe that’s what I do these
days,” says Chakraborty.
“They say, ‘No! It can’t be!’”
Now, the internationally recognized data
mining and analysis expert has found his
true calling as an American father … of
about 700 kids.
Besides his own two teenage sons,
Chakraborty’s family includes his “second
set of kids” who have gone through his
big-data certiﬁcation program in the past
A few years after joining OSU as a quan-
titative marketing professor in 1991, he
worked with Dr. Marilyn Kletke, now-
retired professor in the Management
Science and Information Systems pro-
gram, to launch a graduate-level data
mining certiﬁcation program, combining
aspects of engineering, statistics and
business with other disciplines.
The core level certiﬁcate requires students
to complete 12 credit hours of special-
ized courses, with the expert level cer-
tiﬁcate requiring up to 21 credit hours.
Chakraborty said few students are able to
manage the master’s degree requirements
and instead complete the 21-credit-hour
level of expert certiﬁcation.
Sponsored by business analytics soft-
ware provider SAS, the stand-alone grad-
uate certiﬁcate program attracts students
undertaking master’s programs in various
“If you like playing with data, if you like
solving a problem, if you have intellectual
curiosity and you have some bent of math-
ematical aptitude, this is the best ﬁeld you
can think of,” Chakraborty says.
FILLING THE GAP
As technology throws more and more
data at business people, the issue of how
to make sense of everything has grown.
In a 2011 study, research and consulting
ﬁrm McKinsey and Co. discovered a
yawning gap in that arena of the business
First, the ﬁrm identiﬁed a U.S. shortage
of 140,000 to 190,000 people with deep,
deep analytical expertise. Second, it
identiﬁed a massive need for 1.5 million
people who are not as deeply trained but
know how to use data tools and tech-
niques to make better decisions — the gap
Chakraborty’s program helps ﬁll.
“We can boast as much as we want,
saying we have the best program,” says
Chakraborty. “But let’s talk about valida-
tion from the marketplace.”
In the past 10 years, his students have
presented 160 papers and posters at var-
ious conferences. Twenty-one OSU stu-
dents have been honored with best paper
or best poster awards. In fact, OSU stu-
dents have nailed 12 top prizes in the
SAS-sponsored Analytics Shootout, one
of the largest analytics conferences in
“We have been on the stage every year,”
says Chakraborty. “There is not another
university in the U.S. that has that
Among numerous projects, Chakraborty’s
students are working on a massive data
problem for a Las Vegas casino company.
For three of the company’s properties over
about three years, students are examining
what promotional materials work and
which eorts waste time and money.
The work involves millions and millions of
records and hundreds of columns.
“It’s something that is hard to manipu-
late and to make sense of. But, actually,
if you ask me, that’s not a very big data
problem,” Chakraborty says.
“It’s not, because a truly big data problem
could be billions of records. But from
a teaching point of view, a few million
“You go from a mess of something to
something managers can understand and
take away and put into action,” he says.
“And that’s really what we try to teach in
His students are also working on a social
media eort, examining Twitter and
social media to see how public sentiment
shifted as the Ebola story evolved. They’ve
found people shifted from feeling scared
to mad to learning about preventative
The students examine the data and create
“sentiment mining models” to track peo-
ple’s sentiments about the topic.
BY SONYA COLBERG
engage@spears summer 2015