What this book covers

Chapter 1, R Fundamentals for Machine Learning, introduces you to the problem that will be solved in this book and covers the basics of getting and running R for the subsequent chapters.

Chapter 2, Predicting Failures of Banks - Data Collection, covers the main problems that arise when gathering data and how to structure data to obtain relevant features or variables to develop your first predictive model. 

Chapter 3, Predicting Failures of Banks - Descriptive Analysis, shows how to observe and describe data, how to deal with highly unbalanced data, and also how to deal with missing values in variables.

Chapter 4, Predicting Failures of Banks - Univariate Analysis, covers the analysis and measurement of the individual predictive power of variables and their relationship to the target variable. Additionally, as the number of variables is high, some techniques to reduce the number of variables are also included in this chapter.

Chapter 5, Predicting Failures of Banks - Multivariate Analysis, demonstrates the implementation of different machine learning algorithms. Logistic regression, regularized methods, gradient boosting, neural networks, and Support Vector Machines (SVM) are briefly explained and implemented, to try to obtain an accurate model for predicting bank failures. This chapter also includes some basic guidelines on combining the results of different models to improve the accuracy of our model, and how to generate models in an automatic and visual way.

Chapter 6, Visualizing Economic Problems in Countries, covers the evolution of the financial crisis into a sovereign debt crisis, which shook even the foundations and solvency of the European Union. This chapter shows how macroeconomic imbalances in different countries can be measured. Specifically, this chapter will help you to understand clustering analysis, unsupervised models in nature, and how these techniques can help with even supervised problems.

Chapter 7, Sovereign Crisis - NLP and Topic Modeling, introduces the concept of text mining and topic extraction. This chapter shows that text mining can be very useful to collect information in qualitative reports.