Harnessing the power of "favorites" lists for recommendation systems
Date
2010-01-08T16:12:34Z
Authors
Khezrzadeh, Maryam
Journal Title
Journal ISSN
Volume Title
Publisher
Abstract
This thesis proposes a novel recommendation approach to take advantage of the
information available in user-created lists. Our approach assumes associations among
any two items appearing in a list together. We consider two different ways to calculate the strength of item-item associations: frequency of co-occurrence, and sum
of Bayesian ratings (SBR) of all lists containing the item pair. The latter takes into
consideration not only the number of lists the items have co-appeared in, but also
the quality of the lists. We collected a data set of user ratings for books along with
Listmania lists on Amazon.com using Amazon Web Services (AWS). Our method
shows superior performance to existing user-based and item-based collaborative
filtering approaches according to the resulted Mean Absolute Error (MAE), coverage,
precision and recall.
Description
Keywords
Recommendation system, Association analysis, Amazon, Frequency, Bayesian rating, Collaborative filtering, CIRC