Recommendation Engines
Recommendation Engines
are a type of software that let users rate items, so that other users may learn from the evaluation of th eothers.
Typology
From http://www.readwriteweb.com/archives/recommendation_engines.php
The main approaches fall into the following categories:
- Personalized recommendation - recommend things based on the individual's past behavior
- Social recommendation - recommend things based on the past behavior of similar users
- Item recommendation - recommend things based on the thing itself
- A combination of the three approaches above
Description
Excerpt 1
From the Read/Write blog at http://www.readwriteweb.com/archives/recommendation_engines.php
"A good recommendation engine can make a difference not just for Netflix, but for any online business. This is because there are two fundamental activities online - Search and Browse. When a consumer knows exactly what she is looking for, she searches for it. But when she is not looking for anything specific, she browses. It is the browsing that holds the golden opportunity for a recommendation system, because the user is not focused on finding a specific thing - she is open to suggestions.
During browsing, the user's attention (and their money) is up for grabs. By showing the user something compelling, a web site maximizes the likelihood of a transaction. So if a web site can increase the chances of giving users good recommendations, it makes more money. Obviously this is a difficult problem, but the incentive to solve it is very big." (http://www.readwriteweb.com/archives/recommendation_engines.php)
Excerpt 2
Here's an excerpt from the New York Times, excerpted by Smart Mobs at http://www.smartmobs.com/archive/2006/01/23/ny_times_on_rec.html
See also Collaborative Filtering
"Mr. Hunt said NetFlix's recommendation system collected more than two million ratings forms from subscribers daily to add to its huge database of users' likes and dislikes. The system assigns different ratings to a movie depending on a particular subscriber's tastes. For example, "Pretty Woman" might get a four- or five-star rating if other people who share a customer's taste in movies rated it highly, while the same film might not appear on another customer's screen at all, presumably because other viewers with that customer's tastes did not rate it highly.
"The most reliable prediction for how much a customer will like a movie is what they thought of other movies," Mr. Hunt said. The company credits the system's ability to make automated yet accurate recommendations as a major factor in its growth from 600,000 subscribers in 2002 to nearly 4 million today.
Similarly, Apple's iTunes online music store features a system of recommending new music as a way of increasing customers' attachment to the site and, presumably, their purchases. Recommendation engines, which grew out of the technology used to serve up personalized ads on Web sites, now typically involve some level of "collaborative filtering" to tailor data automatically to individuals or groups of users.
Some engines use information provided directly by the shopper, while others rely more on assumptions, like offering a matching shirt to a shopper interested in purchasing a tie. And some sites are now taking personalization to another level by improving not only the collection of data but the presentation of it.
Liveplasma.com, an online site for music and, more recently, movies, graphically "maps" shoppers' potential interests. A search for music by Coldplay, for example, brings up a graphical representation of what previous customers of Coldplay music have purchased, presented in clusters of circles of various sizes.
The bigger the circle, the greater the popularity of that band. The circles are clustered into orbits representing groups of customers with similar preferences." (http://www.smartmobs.com/archive/2006/01/23/ny_times_on_rec.html)
More Information
- This review describes Amazon, Pandora and Delicious
- Specific treatment of Music Recommendation