In this chapter, we touched upon the topic of time series analysis and forecasting. Of course, we've only scratched the surface, and there is certainly much more to explore. It is also a very important field for the industry, especially in the finance world, with very active research. For example, we see more and more data scientists trying to build time series forecasting models based on recurrent neural network (https://en.wikipedia.org/wiki/Recurrent_neural_network) algorithms, with great success. We've also demonstrated how Jupyter Notebooks combined with PixieDust and the ecosystem of libraries, such as pandas
, numpy
, and statsmodels,
help accelerate the development of analytics as well as its operationalization into applications that are consumable by the line of business user.
In the next chapter, we will look at another important data science use case: graphs. We'll build a sample application related to flight travel and discuss how and when we should apply graph algorithms...