Dec 6, 2013

Big Data

Determining probability of occurrence of events and the temptation of finding correlations between the events have been the aspirations that statisticians have been following during centuries. To accomplish those aspirations, the underlying assumption was that a sample of a population represents that population, within some standard errors.

Big Data, an evolving phenomenon thanks to the information age, online social networks, and information systems, is going to radically change these calculations. Data on customer behavior are collected through systems such as CRM, ERP, etc. in helping businesses predict market trends. Social networking websites are also invaluable tools following customers' footprints ubiquitously. The mobile industry, rising on the ashes of cameras and computers, helps the Big Data get bigger. The to-be released Google Glass is a product generating data, ubiquitously, and has been generated out of the mere concept. The Big Data also opens concerns and debates over privacy issues.

The radical difference between the conventional statistical methodologies and the emerging, ever-growing Big Data are in the concepts of sampling populations and explaining events. Whereas statistics is unable to study the whole population, and contents itself with samples through precise measurements, the Big Data does that instantaneously, though, with less precision. On the other hand, statistics tries to explain events through clarifying positive/negative correlations and predictor-variable relations. In other words, it tries to find out the Why. The Big Data, due to its nature, is incapable of doing that, but it rather focuses on the How.

The new world of business is not a battle between ideas, but a greater war between how you implement the ideas. You don't fight the Big Data, which sounds absurd, you embrace it to win!

No comments:

Post a Comment