Recom
Syst
56 March 1997/Vol. 40, No. 3 COMMUNICATIONS OF THE ACM
Recommender
Systems
Paul Resnick and Hal R. Varian, Guest Editors
Recommender systems assist and augment this
natural social process. In a typical recommender sys-
tem people provide recommendations as inputs,
which the system then aggregates and directs to
appropriate recipients. In some cases the primary
transformation is in the aggregation; in others the
system’s value lies in its ability to make good matches
between the recommenders and those seeking recom-
mendations.
The developers of the first recommender system,
Tapestry
[1], coined the phrase “collaborative filtering”
and several others have adopted it. We prefer the more
general term “recommender system” for two rea-
sons. First, recommenders may not explictly
collaborate with recipients, who may be
unknown to each other. Second, recom-
mendations may suggest particularly
interesting items, in addition to indicat-
ing those that should be filtered out.
This special section includes descrip-
tions of five recommender systems. A sixth
article analyzes incentives for provision of rec-
ommendations.
Figure 1 places the systems in a technical design
space defined by five dimensions. First, the contents of
an evaluation can be anything from a single bit (rec-
ommended or not) to unstructured textual annota-
tions. Second, recommendations may be entered
explicitly, but several systems gather implicit evalua-
tions: GroupLens monitors users’ reading times;
PHOAKS mines Usenet articles for mentions of
URLs; and Siteseer mines personal bookmark lists.
Third, recommendations may be anonymous, tagged
with the source’s identity, or tagged with a pseudo-
nym. The fourth dimension, and one of the richest
areas for exploration, is how to aggregate evaluations.
GroupLens, PHOAKS, and Siteseer employ variants
on weighted voting. Fab takes that one step further to
combine evaluations with content analysis. Referral-
Web combines suggested links between people to form
longer referral chains. Finally, the (perhaps aggregated)
evaluations may be used in several ways: negative rec-
ommendations may be filtered out, the items may be
sorted according to numeric evaluations, or
evaluations may accompany items in a dis-
play.
Figures 2 and 3 identify dimensions
of the domain space: The kinds of
items being recommended and the
people among whom evaluations are
shared. Consider, first, the domain of
items. The sheer volume is an important
variable: Detailed textual reviews of restau-
rants or movies may be practical, but applying
the same approach to thousands of daily Netnews mes-
sages would not. Ephemeral media such as netnews
(most news servers throw away articles after one or two
weeks) place a premium on gathering and distributing
evaluations quickly, while evaluations for 19th century
books can be gathered at a more leisurely pace. The
last dimension describes the cost structure of choices
people make about the items. Is it very costly to miss
IT IS OFTEN NECESSARY TO MAKE CHOICES WITHOUT SUFFICIENT
personal experience of the alternatives. In everyday life, we rely on
recommendations from other people either by word of mouth, rec-
ommendation letters, movie and book reviews printed in newspapers, or
general surveys such as Zagat’s restaurant guides.
