2009 RecommenderSystemsForSocialBookmarking
- (Bogers, 2009) ⇒ Toine Bogers. (2009). “Recommender Systems for Social Bookmarking." PhD Thesis, Tilburg University.
Subject Headings:
Notes
References
1. Introduction
Scope of this Thesis
In this thesis, we investigate how recommender systems can be applied to the domain of social bookmarking. More specifically, we want to investigate the task of item recommendation, where interesting and relevant items — bookmarks or scientific articles — are retrieved and recommended to the user, based on a variety of information sources about the user and the items. This is a difficult task as we are trying to predict which items out of a very large pool would be relevant given a user’s interests, as represented by the items he or she has added in the past. In our experiments we distinguish between two types of information sources.
1.2 Main Contributions
The main contribution of this thesis is a principled investigation of the usefulness of different algorithms and information sources for recommending relevant items to users of social bookmarking services. We extend a standard class of Collaborative Filtering algorithms with information from the folksonomy and compare the results to other state-of-the-art approaches. We also determine the best way of using item metadata for recommendation, and propose several new and hybrid algorithms. We are among the first to compare content-based recommendation with the more common usage-based approaches for social bookmarking services. Compared to related work, we also take a critical look at different methods for fusing recommendations within the same data sets, and determine the optimal ways of combining content-based filtering with collaborative filtering. We cover two different, popular domains, Web pages and scientific articles, and perform extensive evaluations of our our experimental results on publicly available data sets. Finally, we examine two problems with the growth of social bookmarking website: spam and duplicate content. We show how prevalent these phenomena are, propose methods for automatically detecting them, and examine the influence they might have on the item recommendation task.
2.1 Recommender Systems
Another type of technology designed to combat information overload are recommender systems, which have their origin in the field of information filtering (Hanani et al., 2001). A recommender system is used to identify sets of items that are likely to be of interest to a certain user, using a variety of information sources related to both the user and the content items. In contrast to information filtering, recommender systems actively predict which items the user might be interested in and add them to the information flowing to the user, whereas information filtering technology is aimed at removing items from the information stream (Hanani et al., 2001). Over the past two decades many different recommendation algorithms have been proposed for many different domains.
References
- (Hanani et al., 2001) ⇒ Uri Hanani, Bracha Shapira, and Peretz Shova. (2001). “[http://www.nlp.org.cn/docs/20021211/38/Overview-of-Information-Filtering.pdf%7C
Information Filtering: Overview of Issues, Research and Systems].” In: User Modeling and User-Adapted Interaction, 11(3). doi:10.1023/A:1011196000674,