|
Rong Zheng1, Dennis M. Wilkinson2, and Foster
Provost1
(1) Stern School of Business, NYU and (2) Social Computing Laboratory, HP
Labs
Abstract
This paper demonstrates that "social network collaborative
filtering" (SNCF), wherein user-selected like-minded alters are used
to make predictions, can rival traditional user-to-user collaborative
filtering (CF) in predictive accuracy. Using a unique data set from an
online community where users rated items and also created social networking
links specifically intended to represent like-minded “allies,” we use SNCF
and traditional CF to predict ratings by networked users. We find that SNCF
using generic "friend" alters is moderately worse than the better
CF techniques, but outperforms benchmarks such as by-item or by-user
average rating; generic friends often are not like-minded. However, SNCF
using "ally" alters is competitive with CF. These results are
significant because SNCF is tremendously more computationally efficient
than traditional user-user CF and may be implemented in large-scale web
commerce and social networking communities. It is notoriously difficult to
distinguish the contributions of social influence (where allies influence
users) and "social” selection (where users are simply effective at
selecting like-minded people as their allies). Nonetheless, comparing
similarity over time, we do show no evidence of strong social influence
among allies or friends.
Working paper - under review
· Full paper in PDF format
|
|
|
