Man & Machine Teaming for Insight with SPSS

In 1997, Garry Kasparov, a human, famously lost a chess game to IBM Deep Blue, a machine. There-in started the age of “Man vs Machine” – a hot topic for many boardroom debates and box office movies.  What many may not know, is the events that transpired in a freestyle chess tournament held in 2005.  Neither Man nor Machine took home the title, it was in fact two men and a machine working in cooperation that reigned supreme and took home the title.

In Shyam Sankar‘s TED presentation “The rise of human-computer cooperation“, he so rightly points out that “brute computing force alone can’t solve the world’s problems“.  Least of all, the world’s analytical problems.  Which is why Man and Machine must team to drive greater insight.

One of my favorite topics to discuss with clients is predictive analytics.  The value of predictive insight to any business is not only profitable, but highly trans-formative in the way the business thinks and acts.

Even today, confusion still exists over what predictive analytics is and how it is used by the business – with many still fearful of the “black box” that is going to replace their employees and start making key business decisions for them.

This couldn’t be further from the truth.

Predictive analytics is the process of using advanced mathematical algorithms to solve complex business problems, such as identifying hidden trends and patterns in historical information to identify claims that are most likely to be fraudulent, or analyzing customer buying behaviour to identify marketing segments based on their propensity to purchase specific items in the future.

Predictive analytics cannot be applied by either Man or Machine alone.  Humans lack the ability to scale and compute large volumes of data.  Machines lack our ability to think creatively and approach problem solving in a non-linear fashion.  Together, Man and Machine make the perfect team – with Man setting the goals, formulating the hypotheses, identifying the criteria, and iterating the process, whilst Machine performs the computation and executes routine work to scale.

Sankar points out that to make this team operate efficiently and be successful, you have to design the human into the process, ensuring the interface between Man and Machine is void of “friction”.  Predictive analytics is not a question of finding the right algorithm, but rather the right symbiotic relationship between computation and human.

Which is why clients that have turned to IBM SPSS for predictive analytics are so successful.

SPSS Modeler, IBM’s client for designing the workflow required to derive predictive insight, is geared to reduce the “friction” between Man and Machine.  Using a simple, drag-and-drop interface, Man can set the goals of the predictive model, formulate the hypothesis to test, evaluate the results of the model, and employ the Machine to execute the computational work, all without writing any code.   This seamless, configurable process leads to shorter and more efficient model builds which can be further refined and improved in a short space of time – increasing the performance of the predictive model and the Man-Machine team.

When it comes to successfully harnessing the power of predictive analytics: the better the process, the less the friction, the more successful the outcome.

Sankar‘s full presentation is posted below for reference.