Summary

In this chapter, we've talked about both the context and the technical background of machine learning with time-series. Machine learning algorithms or models can make systematic, repeatable, validated decisions based on data. We explained the main machine learning problems with time-series such as forecasting, classification, regression, segmentation, and anomaly detection. We then reviewed the basics of machine learning as relevant to time-series, and we looked at the history and current uses of machine learning for time-series.

We discussed different types of methods based on the approach and features used. Furthermore, we discussed many algorithms, concentrating on state-of-the-art machine learning approaches.

I will discuss approaches including deep learning or classical models, such as autoregressive and moving averages, in chapters dedicated to them (for example, in chapter 5, Time-Series Forecasting with Moving Averages and Autoregressive Models, and chapter...