Book Image

Applied Supervised Learning with R

By : Karthik Ramasubramanian, Jojo Moolayil
Book Image

Applied Supervised Learning with R

By: Karthik Ramasubramanian, Jojo Moolayil

Overview of this book

R provides excellent visualization features that are essential for exploring data before using it in automated learning. Applied Supervised Learning with R helps you cover the complete process of employing R to develop applications using supervised machine learning algorithms for your business needs. The book starts by helping you develop your analytical thinking to create a problem statement using business inputs and domain research. You will then learn different evaluation metrics that compare various algorithms, and later progress to using these metrics to select the best algorithm for your problem. After finalizing the algorithm you want to use, you will study the hyperparameter optimization technique to fine-tune your set of optimal parameters. The book demonstrates how you can add different regularization terms to avoid overfitting your model. By the end of this book, you will have gained the advanced skills you need for modeling a supervised machine learning algorithm that precisely fulfills your business needs.
Table of Contents (12 chapters)
Applied Supervised Learning with R
Preface

Logistic Regression


Logistic regression is the most favorable white-box model used for binary classification. White-box models are defined as models where we have visibility of the entire reasoning used for the prediction. For each prediction made, we can leverage the model's mathematical equation and decode the reasons for the prediction made. There are also a set of classification models that are completely black-box, that is, by no means can we understand the reasoning for the prediction leveraged by the model. In situations where we want to focus on only the end outcome, we should prefer black-box models, as they are more powerful.

Though the name ends with regression, logistic regression is a technique used to predict binary categorical outcomes. We would need a different approach to model for a categorical outcome. This can be done by transforming the outcome into a log of odds ratio or the probability of the event happening.

Let's distill this approach into simpler constructs. Assume...