Spark is very suitable for machine learning projects, such as obtaining a holistic view of business, as it enables us to process huge amounts of data fast and code complicated computations easily. In this section, we will first describe a real business case and then how to prepare Spark computing for our project.
The company IFS sells and distributes thousands of IT products and has a lot of data on marketing, training, team management, promotion, and products. The company wants to understand how various kinds of actions, such as those in marketing and training, affect the sales team's success. In other words, IFS is interested in finding out how much impact marketing, training, or promotion generates separately.
In the past, IFS has done a lot of analytical work, but all of it was completed by individual departments on soloed datasets. That is, they have analytical results about how marketing affects sales from using marketing data alone and how training...