Introduction - Recommender Systems and the Social Web - page 1

Chapter 1
Introduction
1.1 Motivation
Due to the considerable growth of information available online, it has become a constant challenge to
help Internet users to deal with the corresponding information overload. Over the last decade, various
techniques in the areas of information retrieval and filtering have been developed to help users find items
that match their information needs and filter out unrelated information items [Hanani et al., 2001].
In contrast to information retrieval and filtering techniques implemented in search engines, whose
aim is to retrieve the desired information from a large amount of information based on a user query,
recommender systems are today commonly in use on e-commerce platforms. They help online visitors
find relevant information or items to purchase in a personalized way [Jannach et al., 2010; Ricci et al.,
2011a]. In the age of information overload recommender system technologies are of high importance for
the success of large-scale e-commerce sites. Business Insider 1 , for example, names several recommender
system technologies, such as Amazon.com's recommendation engines or Google's news algorithm, among
the 11 most essential algorithms that “make the Internet work”.
When applied in the context of e-commerce, the aim of recommender systems is to provide personalized
recommendations that best suit a customer's taste, preferences, and individual needs [Resnick and Varian,
1997; Adomavicius and Tuzhilin, 2005]. Besides this, a recommender system is supposed to explain the
underlying reasons for its proposals to the user. An example for an explanation would be Amazon.com's
“Customers who bought this item also bought...” label for a recommendation list, which also carries
explanatory information. Explanations for recommendations have increasingly gained in importance over
the last years both in academia and industry because they can significantly influence the user-perceived
quality of such a system [Tintarev and Masthoff, 2007a].
The advantages of using recommender systems are manifold. For example, they can help to build
better relationships with customers, increase the value of e-business, or broaden sales diversity [Nikolaeva
and Sriram, 2006; Dias et al., 2008; Fleder and Hosanagar, 2009]. In practice, recommender systems have
been implemented in different commercial domains such as travel and tourism, entertainment, or topic
sales, as mentioned in [Ricci and Nguyen, 2007], [Jannach and Hegelich, 2009], or [Linden et al., 2003].
With the advent of the Social Web, user generated content has enriched the social dimension of the
Web [Kim et al., 2010b]. New types of Web applications such as Delicious and Flickr 2 have emerged
which emphasize content sharing and collaboration. These so-called Social Web platforms turned users
from passive recipients of information into active and engaged contributors. As a result, the amount of
user contributed information provided by the Social Web poses both new possibilities and challenges for
recommender system research [Freyne et al., 2011].
This thesis focuses on the challenging topic of leveraging these new sources of knowledge to enhance
existing recommender system techniques and introduce new recommendation approaches based on Social
Web data. In particular, we focus on tagging data and propose new ways to leverage user-contributed
tags in recommender systems.
1 http://www.businessinsider.com/internet-algorithms-2011-8
2 http://www.delicious.com , http://www.flickr.com