Chapter 1, A Machine Learning Refresher, presents an overview of machine learning, including basic concepts such as training/test sets, performance measures, supervised and unsupervised learning, machine learning algorithms, and benchmark datasets.
Chapter 2, Getting Started with Ensemble Learning, introduces the concept of ensemble learning, highlighting the problems that it solves as well as the problems that it poses.
Chapter 3, Voting, introduces the most simple ensemble learning technique, voting, while explaining the difference between hard and soft voting. You will learn how to implement a custom classifier, as well as use scikit-learn's implementation of hard/soft voting.
Chapter 4, Stacking, covers meta learning (stacking) a more advanced ensemble learning method. After reading this chapter, you will be able to implement a stacking classifier in Python to use with scikit-learn classifiers.
Chapter 5, Bagging, introduces bootstrap resampling and the first generative ensemble learning technique, bagging. Furthermore, this chapter guides you through the process of implementing the technique in Python, as well as how to use the scikit-learn implementation.
Chapter 6, Boosting, touches on more advanced subjects in ensemble learning. This chapter explains how popular boosting algorithms work and are implemented. Furthermore, it presents XGBoost, a highly successful distributed boosting library.
Chapter 7, Random Forests, goes through the process of creating random decision trees by subsampling the instances and features of a dataset. Moreover, this chapter explains how to utilize an ensemble of random trees to create a random forest. Finally, this chapter presents scikit-learn's implementations and how to use them.
Chapter 8, Clustering, introduces to the possibility of using ensembles for unsupervised learning tasks, such as clustering. Furthermore, the OpenEnsembles Python library is introduced, along with guidance on using it.
Chapter 9, Classifying Fraudulent Transactions, presents an application for the classification of a real-world dataset, using ensemble learning techniques presented in earlier chapters. The dataset concerns fraudulent credit card transactions.
Chapter 10, Predicting Bitcoin Prices, presents an application for the regression of a real-world dataset, using ensemble learning techniques presented in earlier chapters. The dataset concerns the price of the popular cryptocurrency Bitcoin.
Chapter 11, Evaluating Sentiment on Twitter, presents an application for evaluating the sentiment of various tweets using a real-world dataset.
Chapter 12, Recommending Movies with Keras, presents the process of creating a recommender system using ensembles of neural networks.
Chapter 13, Clustering World Happiness, presents the process of using an ensemble learning approach to cluster data from the World Happiness Report 2018.
