Database Reference
In-Depth Information
Chapter 4
Design Considerations for a Social
Network-Based Recommendation
System (SNRS)
Jianming He and Wesley W. Chu
Abstract The effects of homophily among friends have demonstrated their impor-
tance to product marketing. However, it has rarely been considered in recommender
systems. In this chapter, we propose a new paradigm of recommender systems
which can significantly improve performance by utilizing information in social
networks including user preference, item likability, and homophily. A probabilistic
model, named SNRS, is developed to make personalized recommendations from
such information. We extract data from a real online social network, and our
analysis of this large dataset reveals that friends have a tendency to select the
same items and give similar ratings. Experimental results from this dataset show
that SNRS not only improves the prediction accuracy of recommender systems, but
also remedies the data sparsity and cold-start issues inherent in collaborative
filtering. Furthermore, we propose to improve the performance of SNRS by apply-
ing semantic filtering of social networks and validate its improvement via a class
project experiment. In this experiment, we demonstrate how relevant friends can be
selected for inference based on the semantics of friend relationships and finer-
grained user ratings. Such technologies can be deployed by most content providers.
Finally, we discuss two trust issues in recommender systems and show how SNRS
can be extended to solve these problems.
4.1
Introduction
In order to overcome information overload, recommender systems have become a
key tool for providing users with personalized recommendations on items such as
movies, music, books, and news. Intrigued by many practical applications, research-
ers have developed algorithms and systems over the last decade. Some of them have
been commercialized by online venders such as Amazon.com and Netflix.com.
These systems predict user preferences (often represented as numeric ratings) for
Search WWH ::
Custom Search