MIDS Capstone Project Fall 2016

autoCaret

Why autoCaret?

It’s easy to believe that machine learning is hard!

  • We fundamentally believe that the best ideas and concepts in machine learning are simple but that current literature can put up walls up against these ideas.

  • People should be empowered to use machine learning via simple tools (especially in R)!

  • These tools should allow for a linear path enabling their users to graduate into using the underlying language(s) used to make them

  • There needs to be a focus on an analyst’s understanding about what is happening, said differently: robust explanation/summarization about the modeling process is best

 

Where does autoCaret fit?

autoCaret slots into the machine learning context right after everyone's favorite first step, data cleaning! 

autoCaret does the following:

  • Automates training binary classifiers

  • Has an easy to GUI via an R Studio Add-in

  • Wraps the caret and caretEnsemble packages

  • Allows the analyst to learn in the language they will progress into

More Information

Last updated:

March 30, 2017