ABSTRACT
Recommendations via Center-Piece Computation . . . . . . 304 11.4 Exploiting Twitter and Wikipedia for News Recommendations 307
11.4.1 The Blogosphere as a Source of Information . . . . . . . . . . . . 308 11.4.2 Using the Real-Time Web for Personalized News
Recommendations . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 310
11.5 Recommender Systems for Tags . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 313 11.5.1 Social Tagging Platforms as Sources of Information . . . . 313 11.5.2 Recommending Correctly Spelled Tags . . . . . . . . . . . . . . . . . 315
11.6 Conclusion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 317
The explosive growth of the Web due to user generated content has enabled a huge and heterogeneous repository of sources of information and documents. Users may be overwhelmed by the volume of available information. Let us consider, for instance, the social network context. Users may face interaction overload due to the high number of messages generated by their friends, or due to the large number of voters or commenters on a particular Web page, and they may face information overload due to the vast quantity of social media available such as shared photos, video, bookmarks, and so on [249, 418]. To overcome these problems, several methods have been proposed in different but related fields such as Information Retrieval, Information Filtering, and Recommender Systems.
