A Data Science Central Community
That's a big question! I started to throw some ideas on my blog:
I'm interested to have your opinion on the subject.
Data mining is already about automating lot's of insight discovery processes. Automating data mining is automating automation. But who would automate data mining? Data miners? And who would automate the automation of data mining? Or the automation of automaton of automation of data mining?
Nice one :)
Data Mining is not about automation but about a 4 step process which requires extensive and specialized knowledge by the practitioner. This 4 step process is:
1)Identification of the Business Problem
2)Creation of the Analytical File
3)Applying the Right Tools to develop the Solution
4)Deploying the Solution and the Framework for Measurement
Obviously as the data mining discipline has evolved, increasing automation has simply provided better tools for more empowerment by data miners. But if anything, the ability to do more and better solutions has simply increased the need for practitioners or human intervention. Companies that de-emphasize the role of human intervention in the data mining process do so at their own peril.
Thanks to all of you for your input, very interesting!