write:Consumers respond to these personalized pages by purchasing products at much higher rates when compared to banner advertisements and other Web-based pro-motions. At Amazon the rec

assumes customers know what they want and how to describe it when form-ing their search query. For these reasons
many e-commerce sites rely on recommendation engines (sometimes called recommender systems). Recommendation engines proactively identify products that have a high probability of being something the consumer might want to buy. Amazon has long been recognized as having one of the best recommendation engines. Each time customers log into the site
they are presented with an assortment of products based on their purchase history
browsing history
product reviews
ratings
and many other factors. In effect
Amazon customizes their e-commerce site for each individual
leading to increased sales. Consumers respond to these personalized pages by purchasing products at much higher rates when compared to banner advertisements and other Web-based pro-motions. At Amazon
the recommendation engine is credited with generating 35% of sales (Arora
 2016).Recommendation FiltersThere are three widely used approaches to creating useful recommendations: content-based filtering
collaborative filtering
and hybrid strategies (Asrar
 2016).Content-Based Filtering Content-based filtering recommends products based on the product features of items the customer has interacted with in the past (Figure 6.9). Inter-actions can include viewing an item
ÒlikingÓ an item
purchasing an item
saving an item to a wish list
and so on. In the simplest sense
content-based filtering uses item similarity to make recommendations. For instance
the Netflix recommendation engine attempts to recom-mend movies that are similar to movies you have already watched (see IT at Work 6.2). Music-streaming site Pandora creates its recommendations or playlists based on the Music Genome Project©
a system that uses approximately 450 different attributes to describe songs. These detailed systems for describing movies and songs enhance NetflixÕs and PandoraÕs positions in highly competitive industries because of their ability to offer superior recommendations to their customers.3. Recommendation: ÒBased on your rating of fruity cocktail umbrella drink you may also like…Ó1. Customer likes fruity cocktail umbrella drink2. Computer searches products for fruity cocktail umbrella drink FIGURE 6.9 Content-based filtering produces recommendations based on similarity of product features.