Throughout this book, we have seen that ML is very powerful, flexible, and useful for determining and highlighting unexpected events and entities that exist in massive datasets. However, the real value of the technology is often its ability to uncover these insights in near-real time, thus making those insights proactive and actionable. In this chapter, we'll discuss how to effectively integrate ML with Alerting (that is, Watcher). To do this, we will cover the following topics:

• Getting an understanding of how ML's results are published to the results indices 
• Review how the default watch for an ML job works
• Learn how to create a custom watch for advanced functionality

This chapter, however, will not be an extensive overview of Alerting (Watcher). To find out more information about Watcher and its functionality and capabilities, please refer to the Alerting documentation at https://www.elastic.co/guide/en/elastic-stack-overview/current/xpack-alerting.html.