Book Image

AWS Certified Machine Learning Specialty: MLS-C01 Certification Guide

By : Somanath Nanda, Weslley Moura
Book Image

AWS Certified Machine Learning Specialty: MLS-C01 Certification Guide

By: Somanath Nanda, Weslley Moura

Overview of this book

The AWS Certified Machine Learning Specialty exam tests your competency to perform machine learning (ML) on AWS infrastructure. This book covers the entire exam syllabus using practical examples to help you with your real-world machine learning projects on AWS. Starting with an introduction to machine learning on AWS, you'll learn the fundamentals of machine learning and explore important AWS services for artificial intelligence (AI). You'll then see how to prepare data for machine learning and discover a wide variety of techniques for data manipulation and transformation for different types of variables. The book also shows you how to handle missing data and outliers and takes you through various machine learning tasks such as classification, regression, clustering, forecasting, anomaly detection, text mining, and image processing, along with the specific ML algorithms you need to know to pass the exam. Finally, you'll explore model evaluation, optimization, and deployment and get to grips with deploying models in a production environment and monitoring them. By the end of this book, you'll have gained knowledge of the key challenges in machine learning and the solutions that AWS has released for each of them, along with the tools, methods, and techniques commonly used in each domain of AWS ML.
Table of Contents (14 chapters)
Section 1: Introduction to Machine Learning
Section 2: Data Engineering and Exploratory Data Analysis
Section 3: Data Modeling

A word about ensemble models

Before we start diving into the algorithms, there is an important modeling concept that you should be aware of, known as ensemble. The term ensemble is used to describe methods that use multiple algorithms to create a model.

For example, instead of creating just one model to predict fraudulent transactions, you could create multiple models that do the same thing and, using a vote sort of system, select the predicted outcome. The following table illustrates this simple example:

Figure 7.2 – An example of a voting system on ensemble methods

The same approach works for regression problems, where, instead of voting, we could average the results of each model and use that as the final outcome.

Voting and averaging are just two examples of ensemble approaches. Other powerful techniques include blending and stacking, where you can create multiple models and use the outcome of each model as features for a main model. Looking...