Subscribe to DSC Newsletter

The Blind Men and the Elephant in Data Science

I have been following trends in Data Science since long, more from a research perspective ~ Trying to analyze what technique fits in where and what are the new developments in new fields and the bunch of new tools that have been cluttering the market. 

However, I soon realized that every field had it's own modifications and techniques as well as jargon. I encountered Financial Analysts talking about Black-Scholes model as if it's something you learned in primary school. Having worked previously in Marketing and Operations, those were the areas I decided to narrow down upon. Marketing was easy since I had a Market Research background and Factor Analysis and Multi-dimensional scaling was something we Marketers learned in primary school. Supply Chain was a bit tricky since they had Operations Research intensive techniques such as multi- echelon inventory optimization.

Now with the advent of Big Data and Complex algorithms such as Random Forests and LASSO, there is a whole new clutter of jargon, tool and techniques. The boundaries that used to be defined with what technique to be in general used to solve what type of problem are fading. And, it is becoming impossible for one person or even one team to understand and know the gamut. It is just as the blind men trying to analyze the elephant and associating it with something they understood such as a rope, a pillar... On the flip side, I have also come across people discrediting each other's work analogous to the elephant story where one blind man thought it was a rope while another thought it was a pillar.

Lately, I have been taking the plunge into implementing some of these techniques trying to make some sense out of it. I've realized there might not be any value in just analyzing the trends anymore. While I am just another blind man trying to decipher the Data Science world in my own way, I'll try to document my journey in future blogs. 

My take on the process is to put on a Mathematician's hat. Distill and Generalize the problems till the tools and techniques fade and only the equations remain. Once you understand the core aspects, associate it with whatever jargon you want to in whatever field.

Views: 729


You need to be a member of AnalyticBridge to add comments!

Join AnalyticBridge

On Data Science Central

© 2021   TechTarget, Inc.   Powered by

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