Page 96 - Multicloud Workshop - Prework

MapReduce

In this example, the value we are looking for is any word in our dataset and the key is
length. What we end up with is a list of the quantity of all of the words of each length in
our dataset. Let us apply the same concept to our retail company.

The value is “Changing Room Scanner” the key is RFID Code. The result is the
frequency each type of clothing was tried on in all of the changing rooms across all of
out stores. It is reasonable to assume that there would be a correlation between
somebody being attracted enough to try something on and being attracted enough to

buy it. But what if we run our top five changing room items against our top five sold
items and find one of them drops right out. We need to find out why people are trying
but not buying. We need to talk to people but who? Let us run a map reduce with the
RFID as value and store location as key. This will give us the stores where the item is
most tried on. Now we send somebody to do some customer follow up and we find that
although it is liked, loose threads make it appear cheap. A quick Quality fix and sales
rocket. The investment in design, instead of being wasted becomes a real money
maker.

© 2016 Engage ESM All Rights Reserved