Research on the use of social trust relationships for collaborative
ltering has shown that trust-based recommendations
can outperform traditional methods in certain
cases. This, in turn, lead to insights that tie trust
to certain more subtle types of similarity between users
which is not captured in the overall similarity measures
normally used for making recommendations. In this
study, we investigate the use these trust-inspired nuanced
similarity measures directly for making recommendations.
After describing previous research that
identied these similarity statistics, we present an experiment
run on two data sets: FilmTrust and Movie-
Lens. Our results show that using a simple measure -
the single largest dierence between users - as a weight
produces signicantly more accurate results than a traditional
collaborative ltering algorithm and in some
cases also outperforms a model-based approach.
The HCIL has a long, rich history of transforming the experience people have with new technologies. From understanding user needs, to developing and evaluating these technologies, the lab's faculty, staff, and students have been leading the way in HCI research and teaching.