ABSTRACT

Machine learning is an approach to statistical analysis whereby a computer detects and then uses the results to predict outcomes from new data. The combination of large amounts of available data becoming cheaply available and open source machine learning algorithms is causing a revolution in many industries, including water management. For example, water utilities can apply machine learning to predict which water or sewer main is likely to fail soon. This chapter introduces the principles of machine learning and the basic modelling process. The learning objectives for this chapter are:

Understand the principles of machine learning

Apply cross-validation to linear regression

Implement a decision tree prediction