Page 36 - summer18
P. 36

BIG DATA ALLOWS BETTER RESPONSES, SOONER


                                             hat individualized approach is at the heart of   Just as Amazon boasts of having a 360-degree
       ‘When you look at data,            TRam’s Smart Campus work, which Baldasare   view of the customer — being able to predict what
                                          hopes may eventually be incorporated into the   an individual will buy before they do it — so too
       you can see longer-term            predictive work the UA is already doing.  can other organizations leverage data to predict
                                             Ram says generalized solutions are no longer   and respond to the evolving needs of those they
        trends and follow what            enough — in any industry.                serve, Ram says.
                                                                                      “We don’t live in a static world. Our
                                             “We live in an era where you shouldn’t be
                                          generalizing about groups of people,” Ram says.   preferences change, our behavior changes, what
       people are thinking and            “You should be personalizing solutions at the   we do changes, our whole life changes,” Ram
                                          individual level.”                       says. “When you look at data, you can see longer-
         respond to that in real             That thinking underlies many of Ram’s   term trends and follow what people are thinking
                                          projects, ranging from her health care research to   and respond to that in real time in terms of
       time in terms of offering          identify high-cost hospital patients at the point of   offering new services and new products. And
                                          admission to her Smart Cities work in Fortaleza,   you can do interventions and see immediate
         new services and new             Brazil, which aims to tackle public transportation   reactions to these interventions and say, ‘This
                                          challenges. These and other research projects   was a good way to do it; this was not.’ So you’re
                        products.’        take on societal challenges to produce solutions   able to make decisions in a lot shorter time —
                                          with policy and social implications by harnessing   better decisions — and then you can measure
                                          the power of big data, large-scale network   whether your decisions are working or not.”
                                          analysis and machine learning techniques.   In the case of freshman retention,
                                             Coming up with targeted solutions is possible   intervention needs to come early to be most
                                          today more than ever before, thanks to what Ram   effective, whether it is an email from an adviser
                                          refers to as the “datafication phenomenon” — our   or perhaps an invitation from the adviser for the
                                          ability to measure and quantify things that we   student to participate in a time management
                                          couldn’t in the past, from students’ interactions   seminar or some other activity that could boost
                                          on campus to the number of steps it takes to walk   social interaction.
                                          to work in the morning. Those data are generated   “We think by doing these interventions by the
                                          continuously from many different sources, often   12th week, which is when students make up their
                                          with a precise time and location associated with   mind, you’re sort of doing what Amazon does
                                          them.                                    — delivering items you didn’t order but will be
                                             “My philosophy in doing prediction modeling   ordering in the future,” Ram says.
                                          is to combine signals from many different
                                          sources to understand the problem,” Ram says.
                                          “We now live in a sensing environment. We
                                          live in an environment where everything is
                                          connected, so our whole world is even more
                                          than an ‘internet of things’ — it is an internet of
                                          people, data and things. It’s all three connected to
                                          each other, and it’s all about understanding those
                                          connections. This is where network science and
                                          machine learning interact and allow us to make
                                          better predictions.”






        34  ARIZONA ALUMNI MAGAZINE
   31   32   33   34   35   36   37   38   39   40   41