A recommendation engine is a tool that recommends products to consumers based on user preferences or being similar to other users when performing big data analysis. It is a tool that is the key ingredient in creating recommender systems. The challenge of creating a product recommendation given large quantities of data with ratings by users is a popular topic since the Netflix Prize competition. We will provide one engine example based on the Recommenderlab package available here for the Statistical Computing with R tool. The steps required to achieve a recommendation when working with such an engine is modeled after the Cross Industry Standard Process for Data Mining (CRISP-DM) and you find more information about it here.
Step 1 Business Understanding
The goal in this example is to create a movie recommendation engine using R in order to find movies that users want to rent based on previous ratings given to movie titles by users.
Step 2 Data Understanding
The freely available data set we are using consists of movie ratings obtained by users for a wide variety of different movies. In order to understand the data better we refer to our article movie ratings.
Recommender engine tutorial
The following video provides more pieces of information about the topic with a tutorial using a Python toolset: