Summary

In this chapter, you have completed a real-world taxi demand prediction application on the Spark platform.

We have shown you how to work with big data in KNIME and how to prototype big data workflows before accessing a remote cluster. You have, therefore, acquired a toolkit for accessing, preprocessing, and analyzing big data, which enables you with the full computational power of a remote cluster while working from the convenient visual environment in KNIME.

You have worked through training, testing, and deploying a demand prediction application using the random forest algorithm, enabling you with the necessary skills of building demand prediction applications in other fields as well, given that demand prediction is one of the classic applications of time series analysis.

In the next chapter, we extend the demand prediction problem into a multivariate case using an LSTM model.