It is easy to know something, but difficult to put that knowledge into practiceChinese proverb
The different recommendation techniques presented in Chapter 2 are used in a variety of contexts, including commercial, industrial and academic applications. This chapter presents a number of ways in which these techniques are implemented in practice.
Many systems that we use on a daily basis offer recommendations to their users: for example groups, jobs and people may be recommended by LinkedIn [LIN 15]; Facebook [FAC 15] recommends friends; systems such as Last.fm [LAS 15] recommend music, and Websites such as Forbes.com [FOR 15] recommend news stories. In this section, we describe the recommendation techniques used by Amazon.com [AMA 15] (for product recommendations) and by Netflix [NET 15] (for movie recommendations).
Recommender systems are widely used by online retailers. Websites such as Amazon.com [AMA 15] (or other similar online retailers) present users with suggestions of products that they may wish to buy.
Recommendation algorithms are widely known due to their use by e-commerce sites, where client interests are used as input in order to generate lists of recommended products. Many applications only use details of products that clients have purchased and evaluated explicitly in order to represent these interests, but other attributes may also be included, such as lists of consulted products, demographic ...