Feature Engineering in R

Overview

Learn the principles of feature engineering for machine learning models and how to implement them using the R tidymodels framework.

In this course, you’ll learn about feature engineering, which is at the heart of many times of machine learning models. As the performance of any model is a direct consequence of the features it’s fed, feature engineering places domain knowledge at the center of the process. You’ll become acquainted with principles of sound feature engineering, helping to reduce the number of variables where possible, making learning algorithms run faster, improving interpretability, and preventing overfitting.

You will learn how to implement feature engineering techniques using the R tidymodels framework, emphasizing the recipe package that will allow you to create, extract, transform, and select the best features for your model.

When faced with a new dataset, you will be able to identify and select relevant features and disregard non-informative ones to make your model run faster without sacrificing accuracy. You will also become comfortable applying transformations and creating new features to make your models more efficient, interpretable, and accurate!

Syllabus

Introducing Feature Engineering

Raw data does not always come in its best shape for analysis. In this opening chapter, you will get a first look at how to transform and create features that enhance your model's performance and interpretability.

Transforming Features

In this chapter, you’ll learn that, beyond manually transforming features, you can leverage tools from the tidyverse to engineer new variables programmatically. You’ll explore how this approach improves your models' reproducibility and is especially useful when handling datasets with many features.

Extracting Features

You’ll now learn how models often benefit from reducing dimensionality and extracting features from high-dimensional data, including converting text data into numeric values, encoding categorical data, and ranking the predictive power of variables. You’ll explore methods including principal component analysis, kernel principal component analysis, numerical extraction from text, categorical encodings, and variable importance scores.

Selecting Features

You’ll wrap up the course by learning about feature engineering and machine learning techniques. You’ll begin by focusing on the problems associated with using all available features in a model and the importance of identifying irrelevant and redundant features and learning to remove these features using embedded methods such as lasso and elastic-net. Next, you’ll explore shrinkage methods such as lasso, ridge, and elastic-net, which can be used to regularize feature weights or select features by setting coefficients to zero. Finally, you’ll finish by focusing on creating an end-to-end feature engineering workflow and reviewing and practicing the previously learned concepts and functions in a small project.