/Machine-Learning-Methods-in-R

An almost comprehensive project comparing and analyzing basic machine learning methods on a moderate-size dataset with 12 continuous inputs and two discrete inputs.

Primary LanguageHTML

Machine-Learning-Methods-in-R

An almost comprehensive project comparing and analyzing basic machine learning methods on a moderate-size dataset with 12 continuous inputs and two discrete inputs.

Requirements

  • R programming language
  • Tidyverse library
  • Caret library (included in Tidyverse library)
  • Other libraries specific to different parts of project

Machine learning methods used

  • Simple Linear Regression (with basis functions)
  • Regularized Regression with Elastic net (pair and triplet interactions)
  • Logistic Regression (Generalized Linear Regression)
  • Neural Network (single hidden layer, regression and classification)
  • Random Forest (regression and classification)
  • Gradient Boosted Tree (regression and classification)
  • Support Vector Machines (regression and classification)
  • Multivariate Additive Regression Splines (MARS) (regression and classification)
  • Partial Least Squares