Book Image

The Data Science Workshop - Second Edition

By : Anthony So, Thomas V. Joseph, Robert Thas John, Andrew Worsley, Dr. Samuel Asare
5 (1)
Book Image

The Data Science Workshop - Second Edition

5 (1)
By: Anthony So, Thomas V. Joseph, Robert Thas John, Andrew Worsley, Dr. Samuel Asare

Overview of this book

Where there’s data, there’s insight. With so much data being generated, there is immense scope to extract meaningful information that’ll boost business productivity and profitability. By learning to convert raw data into game-changing insights, you’ll open new career paths and opportunities. The Data Science Workshop begins by introducing different types of projects and showing you how to incorporate machine learning algorithms in them. You’ll learn to select a relevant metric and even assess the performance of your model. To tune the hyperparameters of an algorithm and improve its accuracy, you’ll get hands-on with approaches such as grid search and random search. Next, you’ll learn dimensionality reduction techniques to easily handle many variables at once, before exploring how to use model ensembling techniques and create new features to enhance model performance. In a bid to help you automatically create new features that improve your model, the book demonstrates how to use the automated feature engineering tool. You’ll also understand how to use the orchestration and scheduling workflow to deploy machine learning models in batch. By the end of this book, you’ll have the skills to start working on data science projects confidently. By the end of this book, you’ll have the skills to start working on data science projects confidently.
Table of Contents (16 chapters)
Preface
12
12. Feature Engineering

Ensemble Learning

Ensemble learning, as the name denotes, is a method that combines several machine learning models to generate a superior model, thereby decreasing variability/variance and bias, and boosting performance.

Before we explore what ensemble learning is, let's look at the concepts of bias and variance with the help of the classical bias-variance quadrant, as shown here:

Figure 15.1: Bias-variance quadrant

Variance

Variance is the measure of how spread out data is. In the context of machine learning, models with high variance imply that the predictions generated on the same test set will differ considerably when different training sets are used to fit the model. The underlying reason for high variability could be attributed to the model being attuned to specific nuances of training data rather than generalizing the relationship between input and output. Ideally, we want every machine learning model to have low variance.

Bias

Bias...