Scaling Up Prediction to Terabyte Click Logs
In the previous chapter, we developed an ad click-through predictor using a logistic regression classifier. We proved that the algorithm is highly scalable by training efficiently on up to 1 million click log samples. In this chapter, we will further boost the scalability of the ad click-through predictor by utilizing a powerful parallel computing (or, more specifically, distributed computing) tool called Apache Spark.
This chapter will demystify how Apache Spark is used to scale up learning on massive data, as opposed to limiting model learning to one single machine. We will also use
PySpark, which is the Python API, to explore click log data, to develop classification solutions based on the entire ...