Frozen versus liquid analytics - or the competitive advantage of data science

In this article, I further emphasize the difference between data scientists and other analytic practitioners. I wrote last week that statisticians are to data scientists what astronomers are to physicists: there's some overlap, but less than most people think. Here, I elaborate on this theme.

Other disciplines such as data mining and machine learning (and of course statistics) produce frozen statistics: analyses produced by an employee or consultant, resulting in insights delivered to managers, or integrated in tools developed by software engineers. Synonymous include static or solid analytics.

To the contrary, data science is also about liquid analytics. Synonymous include adaptive or interactive analytics. Extreme liquid analytics is called real-time analytics.

The two components of data science

  • Frozen analytics to create and prototype rule and scoring system, using cross-validation, training sets, sampling and algorithms like traditional machine learning algorithms.
  • Liquid analytics. That's the part that automatically updates and refines the training sets, rules, inferences, confidence intervals and predictions, every day, as mutating data keeps pouring non-stop in the databases (be it NoSQL or not). While this (most of the time) still ends up being coded in production mode by software engineers or developers, the framework and logical architecture is designed by data scientists. Because of this, data science is to data floods what statistical science is to frozen data.

Many tools and software packages are designed to handle frozen analytics. Few can handle liquid analytics, though if you know your tool well (and integrate it with other tools) or write your own code, you'll be able to work the liquid analytics part. I did it with a combination of HTML, SAS, Perl/CGI, sendmail, cronjobs, UNIX commands, shell scripts, Sybase/SQL (and an API offered to users), back in 1997: read the section My first years in the corporate world in My Data Science Journey.

Also, not all data scientists work on liquid analytics. Many actually assigned the data science job title to themselves, while still carrying on traditional frozen analytics: they are to data science what quacks are to modern medecine. Finally, without liquid analytics, it's very hard to produce robust, automated, scalable data science, your best option being then light analytics.

Related articles

Views: 2189


You need to be a member of Data Science Central to add comments!

Join Data Science Central

© 2021   TechTarget, Inc.   Powered by

Badges  |  Report an Issue  |  Privacy Policy  |  Terms of Service