Bayesian MixedEects Models for Recommender Systems

Recommender systems have emerged as an important
application area and have been the focus of considerable
recent academic and commercial interest The  spe
cial issue of the Communications of the ACM   con
tains some key papers Other important contributions
include 
       
        
  and   In
addition many online retailers are using this technol
ogy to recommend new items to their customers based
on what they have bought in the past
Currently most recommender systems are either
contentbased or collaborative depending on the type of
information that the system uses to recommend items
to a user
Contentbased lters try to recommend items that
are similar to those that a given user has liked in the
past These approaches are usually based on techniques
from information retrieval or machine learning For
example text documents are recommended based on a 数据挖掘研究院
comparison between their content and a users prole
The prole is built up by analyzing the content of the
items that the user has already rated Thus the system
tries to recommend items that are similar to those that
a user has liked in the past These types of systems require both that a user has rated some items and that
there exists a way to derive explicit contentoriented
features from the item
Alternatively collaborative ltering systems try to
identify other users with tastes similar to the current
user and recommend items that those users have liked
This approach appeals to the notion that when we
are looking for information we often seek the advice
of friends with similar tastes or other people whose
judgment we trust We should be able to make use of
what others have already found and evaluated Again
this approach requires that a new user rate some items
but it typically does not require that items or users be

数据挖掘研究院

  数据挖掘实验室