Skip to main content
main-content

Über dieses Buch

​There is an increasing demand for recommender systems due to the information overload users are facing on the Web. The goal of a recommender system is to provide personalized recommendations of products or services to users. With the advent of the Social Web, user-generated content has enriched the social dimension of the Web. As user-provided content data also tells us something about the user, one can learn the user’s individual preferences from the Social Web. This opens up completely new opportunities and challenges for recommender systems research. Fatih Gedikli deals with the question of how user-provided tagging data can be used to build better recommender systems. A tag recommender algorithm is proposed which recommends tags for users to annotate their favorite online resources. The author also proposes algorithms which exploit the user-provided tagging data and produce more accurate recommendations. On the basis of this idea, he shows how tags can be used to explain to the user the automatically generated recommendations in a clear and intuitively understandable form. With his book, Fatih Gedikli gives us an outlook on the next generation of recommendation systems in the Social Web sphere.

Inhaltsverzeichnis

Frontmatter

Chapter 1. Introduction

Abstract
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].
Fatih Gedikli

Chapter 2. Preliminaries

Abstract
This chapter is organized as follows: First we present the two basic filtering techniques for recommender systems: collaborative and content-based filtering. Afterwards, in Section 2.2, we discuss how recommender systems can be compared according to different quality aspects that are relevant for the application. We describe the different types of evaluation procedures commonly found in the recommender system literature and their respective application domains. Finally, in Section 2.3, we present current recommendation approaches based on Social Web data. We focus on tagging data and identify the different approaches to leverage tagging data in recommender systems.
Fatih Gedikli

Chapter 3. LocalRank – A graph-based tag recommender

Abstract
Tag recommenders are designed to help the online user in the tagging process and suggest appropriate tags for resources with the purpose to increase the tagging quality [Jäschke et al., 2008]. In recent years, different algorithms have been proposed to generate tag recommendations given the ternary relationships between users, resources, and tags, see, for example, [Rendle et al., 2009; Rendle and Schmidt-Thie, 2010] or [Gemmell et al., 2010]. Many of these algorithms, however, suffer from scalability and performance problems, including the popular FolkRank algorithm [Hotho et al., 2006]. For example, even when using only a small excerpt of a commonly used social bookmarking data set, FolkRank requires about 20 seconds on a typical desktop PC (AMD Athlon II Dual Core, 2.9Ghz, 8GB Ram) to compute a single recommendation list.
Fatih Gedikli

Chapter 4. Improving recommendation accuracy based on item-specific tag preferences

Abstract
Recent research has indicated that “attaching feelings to tags” is experienced by users as a valuable means to express which features of an item they particularly like or dislike [Vig et al., 2010]. When following such an approach, users would therefore not only add tags to an item as in usual Web 2.0 applications, but also attach a preference (affect) to the tag itself, expressing, for example, whether or not they liked a certain actor in a given movie. In this chapter, we show how this additional preference data can be exploited by a recommender system to make more accurate predictions.
Fatih Gedikli

Chapter 5. Evaluation of explanation interfaces in the form of tag clouds

Abstract
Current research has shown the important role of explanation facilities in recommender systems based on the observation that explanations can significantly influence the user-perceived quality of such a system [Tintarev and Masthoff, 2012]. In this chapter, we present and evaluate explanation interfaces in the form of tag clouds, which are a frequently used visualization and interaction technique on the Web [Lohmann et al., 2009]. We report the result of a user study in which we compare the performance of two new explanation methods based on personalized and non-personalized tag clouds with a previous explanation approach. Overall, the results show that explanations based on tag clouds are not only well-accepted by the users but can also help to improve the efficiency and effectiveness of the explanation process.
Fatih Gedikli

Chapter 6. An analysis of the effects of using different explanation styles

Abstract
When explaining recommendations to the customers, one of the main challenges is how to select the appropriate presentation interface for explanations. Good explanations not only have to be easily understandable by the end user but should also help the user to make good decisions [Bilgic and Mooney, 2005; Tintarev and Masthoff, 2012]. This chapter addresses the question of how explanations can be communicated to the user in the best possible way. To that purpose, we analyze ten different explanation interfaces with respect to six evaluation factors in a user study.
Fatih Gedikli

Chapter 7. Summary and perspectives

Abstract
The goal of recommender systems is to provide personalized recommendations of products or services to users facing the problem of information overload on the Web [Adomavicius and Tuzhilin, 2005]. They provide personalized recommendations that best suit a customer’s taste, preferences, and individual needs. Especially on large-scale Web sites where millions of items such as books or movies are offered to the users, recommender system technologies play an increasingly important role. One of their main advantages is that they reduce a user’s decision-making effort [Felfernig et al., 2011; Ricci et al., 2011b]. However, recommender systems are also of high importance from the service provider or system perspective. For instance, they can convince a customer to buy something or develop trust in the system as a whole which ensures customer loyalty and repeat sales gains [Jannach et al., 2010; Ricci et al., 2011a].
Fatih Gedikli

Backmatter

Weitere Informationen

Premium Partner

    Bildnachweise