🎉 🎉 RAPIDMINER 9.10 IS OUT!!! 🎉🎉

Download the latest version helping analytics teams accelerate time-to-value for streaming and IIOT use cases.

CLICK HERE TO DOWNLOAD

Where data scientists cannot be automated

tonyboy9tonyboy9 Member Posts: 94 Contributor II
This story appeared recently in Towards Data Science. Since I've been learning RapidMiner for only one year, someone more experienced please give me your opinions.                                       

https://towardsdatascience.com/5-examples-where-data-scientists-cant-be-automated-c3d82c518d37

Best Answer

  • kaymankayman Member Posts: 652   Unicorn
    Solution Accepted
    Yeah, I read it too and thought it was spot on.

    AutoML will not replace (all of) us, it will mainly make our jobs easier and less repetitive. A good scientist will still have to be able to translate a business requirement in a workflow, will still have to understand where the data is coming from and where it has to go in whatever format, and will still have to spend more time than wanted on cleaning the garbage out of the data.

    If AutoML then supports us in the next step to decide which models work best, fine. A seasoned data scientist does the same in the end without thinking, so it just saves time in the end.

    Now, this goes for decent scientists. these will not be impacted negatively. It's the mediocre or bad ones that might be replaced by AML, and maybe that's not too bad for the whole of the business
    mschmitzlionelderkrikoryyhuang
Sign In or Register to comment.