Overview of available machine learning frameworks in R, which are ordered from cutting-edge to outdated:
tidymodels: collection of packages for modeling and machine learning using tidyverse principle covering the whole workflow (data pre-processing to evaluating model performance). Should also make it simple to switch to between different models. Recommended e-learning course (~4h) is offered by DataCamp Modeling with tidymodels in R.
mlr3: R package developed by statisticians at TU Dortmund. Should make it easy to develop different machine learning models. Maybe an alternative to tidymodels approach?
caret: R package for machine learning. Outdated, as main developer Max Kuhn now works for R Studio for improving the tidymodels framework mentioned above.
Books
Max Kuhn & Julia Silge (2021): Tidy Modeling with R (free HTML version)
Becker et al. 2021: mlr3 book (free HTML version)
Chester Ismay & Albert Y. Kim (2021): Statistical Inference via Data Science - A ModernDive into R and the Tidyverse (free HTML version)
Bradley Boehmke & Brandon Greenwell (2020): Hands-On Machine Learning with R (free HTML version)
Hefin Ioan Rhys (2020): Machine Learning with R, the tidyverse, and mlr (partly free HTML version)
General Book Recommendation for …
“… R users who want to improve their programming skills and understanding of the language. It should also be useful for programmers coming to R from other languages,”
— Hadley Wickam (2021): Advanced R (2nd edition) (free HTML version)
E-Learning plattform Datacamp: offers more than 37 Machine Learning courses in R. At KWB we have a premium company account. If you are working for KWB and interested please follow the Learning R on DataCamp workflow.