Database Reference
In-Depth Information
Summary
In this chapter, we connected some of the dots between online machine learning and
streaming data analysis. We introduced the Spark Streaming library and API for continuous
processing of data streams based on familiar RDD functionality and worked through ex-
amples of streaming analytics applications that illustrate this functionality.
Finally, we used MLlib's streaming regression model in a streaming application that in-
volves computing and comparing model performance on a stream of input feature vectors.